the impact of a tqm intervention on work attitudes: a longitudinal

THE IMPACT OF A TQM INTERVENTION ON

WORK ATTITUDES: A LONGITUDINAL

CASE STUDY

A dissertation submitted for the degree of PhD (Econ)

Faculty of Economics London School of Economics and Political Science

Jacqueline Coyle-Shapiro 1996

THESIS ABSTRACT

Total Quality Management (TQM) has been heralded as a new way of managing

organizations. While there are widespread endeavours by organizations to implement

TQM, a visible lag exists between the adoption of TQM and a systematic evaluation of this

phenomenon. The thesis, therefore, addresses a fundamental question in TQM; what is the

impact, if any, of a TQM intervention on employee work attitudes?

This 'before and after study' examines the impact of a 'soft' TQM intervention on two key

elements of TQM: teamwork and continuous improvement. A questionnaire was

completed by respondents six months prior to and nine months after the launch of the

intervention. The starting point in the evaluation is the development of theoretical models

containing hypothesized antecedents of teamwork and continuous improvement which are

empirically tested on the data. The intervention is then evaluated on the basis of its direct

and indirect effects on the two key elements of TQM. In addition, the impact of the

intervention is assessed both at the individual and the organizational level.

At the individual level, the intervention was found to have a significant effect on team

orientation as well as on a number of dimensions of continuous improvement, including

general orientation to quality, improvement as part of the job and intrinsic motivation.

However, a significant overall improvement at the organizational level was not evidenced

in any of these variables. This raises the possibility that a longer time lag may be required

for the individual level effects to develop into an overall organizational improvement.

Additional important findings emerged from this evaluation. First, a consistent finding

throughout is the importance of supervisory behaviour in affecting employee attitudes.

Second, employee assessment of the intervention is a more significant predictor of

subsequent changes than employee participation in the intervention per se. Finally, the

prior experience and attitudes of individuals have a significant effect on how the

intervention is assessed, which subsequently affects changes in attitudes, highlighting the

fact that organizational change interventions do not occur in a vacuum.

TABLE OF CONTENTS

Abstract

List of Tables/Figures vii

Acknowledgements

Chapter 1 Total Quality Management: Theory and Research 1

1.1 Introduction 1

1.2 Why investigate TQM? 3

1.3 What is TQM? 6 1.3.1 Core elements of TQM 7

1.4 A critical analysis of the core elements 10 1.4.1 Theoretical basis 11 1.4.2 Teamwork 12 1.4.3 Continuous improvement 14 1.4.4 Change 18

1.5 Empirical research 21

1.6 Research objectives 24

1.7 Structure of the thesis 26

Chapter 2 The Context 27

2.1 Introduction 27

2.2 The Organization 28 2.2.1 The Site 29 2.2.2 Major changes at the site 30

2.3 Rationale behind WTTW 33

2.4 The TQM intervention 35

2.5 The process of implementation 39

2.6 Summary 42

ii

Chapter 3 Research Methodology 43

3.1 Introduction 43

3.2 Research questions 44

3.3 Research approaches: choice? 48 3.3.1 Evaluation: organizational change interventions 50 3.3.2 Ideal research design 51

3.4 Research design adopted 54 3.4.1 Case study approach 55 3.4.2 Longitudinal research 56 3.4.3 A comparison site 58 3.4.4 The questionnaires 59 3.4.5 Data collection 61 3.4.6 Sample characteristics 62

3.5 Ideal vs. actual research design: limitations 63

3.6 Data analysis 65

3.7 Conclusions 66

Chapter 4 The Antecedents of Team Orientation 68

4.1 Introduction 68

4.2 Team orientation 72 4.2.1 Definition of team orientation 72 4.2.2 The antecedents of team orientation 74

4.3 Measures 79

4.4 Analysis procedures 82

4.5 An empirical testing of the antecedents of team orientation 83 4.5.1 Descriptive statistics 83 4.5.2 Results 85

4.6 Discussion 87

4.7 Conclusions 91

iii

Chapter 5 The Impact of a TQM Intervention on Team Orientation 92

5.1 Introduction 92

5.2 Measures of the intervention 94 5.2.1 Factor analysis of items measuring the intervention 96


5.4 Descriptive statistics of the TQM intervention 99

5.5 Results 101

5.6 Discussion 104

5.7 Conclusions 112

Chapter 6 Commitment to Improvement and Intrinsic Motivation: The Same Constructs? 113

6.1 Introduction 113

6.2 Commitment to improvement 115 6.2.1 Definition of commitment to improvement 116 6.2.2 Factor analysis of commitment to improvement 118

6.3 Intrinsic motivation: A different construct? 119

6.4 The antecedents of commitment to improvement 123

6.5 Measures 128

6.6 Analysis procedures 131 6.6.1 Measurement of change using retrospective data 131

6.7 Results 134 6.7.1 Site 1 results 134 6.7.2 Site 2 results 137

6.8 Discussion 139

6.9 Conclusions 148

iv

Chapter 7 The Impact of a TQM Intervention on Commitment to Improvement and Intrinsic Motivation 149


7.2 Evaluation model 150

7.3 Descriptive statistics 152

7.4 Results 154

7.5 Discussion 160

7.6 Conclusions 168

Chapter 8 Predictors of Employee Participation in a TQM Intervention 170


8.2 An hypothesized model of employee participation in a TQM intervention 175

8.3 Additional model and analysis procedures 181


8.5 Results 183

8.6 Discussion 185

8.7 Conclusions 189

Chapter 9 Predictors of Employee Assessment of a TQM Intervention 191


9.2 An hypothesized model of employee assessment of a TQM intervention 196

9.3 Measures 199


9.5 Results 201

9.6 Discussion 204

9.7 Conclusions 211

Chapter 10 Supervisory Behaviour 213



10.3 An hypothesized model of supervisor participative style and commitment to quality 223

10.4 Measures and analysis procedures 227

10.5 Results 232

10.6 Discussion 233

10.7 Conclusions 240

Chapter 11 An Overall Assessment of the TQM Intervention 242


11.2 Criteria of assessment 249 11.2.1 Elements of TQM 250 11.2.2 Performance outcomes 253 11.2.3 Other outcomes 259 11.2.4 Goal achievement 262 11.2.5 A TQM view 267

11.3 Discussion 268

11.4 Conclusions 270

Chapter 12 Conclusions 272


12.2 Main findings 273

12.3 The extent of change 279

12.4 Greater change? 283

12.5 Limitations of the research 284

12.6 Future research 285

Appendix 288

Bibliography 382

vi

LIST OF TABLES

Table 3.1 Sample characteristics of sitel 63

Table 4.1 Independent t-tests and paired sample t-tests for the individual site samples (employees) 84

Table 4.2 The impact of the antecedent and control variables on team orientation at time 2 86

Table 4.3 The impact of changes in the antecedent variables (and the control variables) on changes in team orientation 87

Table 4.4 The impact of the antecedent and control variables on team orientation at time 1 89

Table 5.1 Factor analysis of the items measuring the intervention 97

Table 5.2 Descriptive statistics of the TQM intervention variables 99

Table 5.3 Direct, indirect and total effects of the TQM intervention on variables in the model at Time 2 102

Table 5.4 Direct, indirect and total effects of TQM intervention on variables in the model- change over time (T2-T1) 103

Table 6.1 Factor analysis of items measuring commitment to improvement 118

Table 6.2 Factor analysis of items measuring intrinsic motivation and commitment to improvement 121

Table 6.3 Means and standard deviations of general orientation to quality, improvement as part of the job and conceptually similar measures 132

Table 6.4 The impact of the antecedent variables on the three dependent variables at time 2 - site 1 134

Table 6.5 The impact of changes in the antecedent variables on changes in the three dependent variables - site 1 135

Table 6.6 The impact of the antecedent variables on the three dependent variables at time 2 - site 2 137

vii

Table 6.7 The impact of changes in the antecedent variables on changes in the dependent variables - site 2 138

Table 6.8 Significant antecedents of general orientation to quality, improvement as part of the job and intrinsic motivation at site 1 and site 2 at time 2 143

Table 6.9 Significant antecedents of changes in general orientation to quality, improvement as part of the job and intrinsic motivation at site 1 and site 2 144

Table 6.10 Summary of significant antecedents of variations and changes in general orientation to quality, improvement as part of the job and intrinsic motivation at site 1 and site 2 147

Table 7.1 Paired sample t-tests for employees at site 1 153

Table 7.2 Direct effects of the TQM intervention on the intervening variables in the model at time 2 155

Table 7.3 Direct, indirect and total effects of the TQM intervention on the dependent variables at time 2 156

Table 7.4 Direct effects of the TQM intervention on intervening variables in the model -change over time (T2-T1) 158

Table 7.5 Direct, indirect and total effects of the TQM intervention on the dependent variables -change over time (T2-T1) 159

Table 8.1 Independent t-tests between participants and non-participants in the TQM intervention 182

Table 8.2 Predictors of employee participation in a TQM intervention 184

Table 9.1 Predictors of employee assessment of a TQM intervention (using same predictors as for employee participation) 202

Table 9.2 Predictors of employee assessment of a TQM intervention (full model) 203

Table 10.1 Paired sample t-tests for the group of managers / supervisors at site 1 218

viii

Table 10.2 Independent t-tests between managers / supervisors and employees at site 1 220

Table 10.3 Predictors of supervisor participative style and commitment to quality 232

Table 11.1 Summary table of the impact of the TQM intervention 248

Table 11.2 Predictors of perceived perfoimance improvement 257

Table 11.3 Independent t-tests between participants and non participants 261

LIST OF FIGURES

Figure 3.1 Logic underlying the evaluation of a TQM intervention 45

Figure 4.1 Hypothesized model of the antecedents of team orientation 75

Figure 5.1 Hypothesized evaluation model of the impact of a TQM intervention on team orientation 93

Figure 6.1 Hypothesized model of the antecedents of commitment to improvement 124

Figure 7.1 Hypothesized evaluation model of a TQM intervention on commitment to improvement 151

Figure 8.1 Hypothesized model of employee participation in a TQM intervention 176

Figure 9.1 Full hypothesized model of employee assessment of a TQM intervention 198

Figure 10.1 Hypothesized model of supervisor participative style and commitment to quality 224

Figure 11.1 The role of organizational commitment and supervisor participative style in the process of change 245

ix

ACKNOWLEDGEMENTS

Numerous individuals contributed to this research. First, I would like to express my

gratitude to all the individuals at the two sites for their continued co-operation without

which this research would not have been possible. In particular, the co-operation of the

two General Managers in allowing me to spend considerable time at the sites is

acknowledged. In addition, my appreciation is extended to Brian Colgate who supported

this research from the beginning and proactively pursued access to the two sites.

I would like to extend special thanks to my supervisor Riccardo Peccei who provided

excellent direction and support and from whom I have learned a great deal.

This research has benefited from the constructive comments from my fellow PhD students

in seminars. I am also indebted to the members of the Industrial Relations Department for

their support.

Finally, I would like to express my gratitude to Mark for his unfailing support over the

course of this endeavour.

Chapter 1: Total Quality Management: Theory and Research

1.1 Introduction

Total Quality Management (TQM) has been heralded as a new way of managing

organizations. While it has generated considerable interest from the practitioner

community academic interest has lagged behind. It is not uncommon for a considerable

lag to exist between innovations adopted by organizations and their systematic

evaluation. As such, TQM is no exception. The view presented by Steel and Jennings

(1992) is still widely applicable to TQM research, "there are no systematic and

controlled studies of the TQM process" (p25). However, a few evaluation studies have

recently been conducted (Guest and Peccei, 1994; Peccei and Wood, 1994; Wood and

Peccei, 1995).

The absence of evaluation research may in the most part be due to the nature of TQM.

By its very nature, TQM is difficult to evaluate. There are several reasons for this.

First, TQM, as a concept is ill defined and related to this, it has little theoretical

foundation to guide evaluations. Second, the intended scope of TQM change exceeds

the more typical organizational change interventions (Steel and Jennings, 1992). Third,

the multifaceted nature of TQM makes it difficult to isolate the effects of individual

changes. Finally, any scientific evaluation of TQM would have to adopt a longitudinal

approach and overcome the unique potential problems inherent in undertaking

longitudinal studies.

Overall, relatively little is known about whether TQM works; whether it does indeed

bring about the desired positive changes in attitudes, behaviour and performance as

claimed by its proponents. This thesis sets out to evaluate a TQM intervention and

1

thereby makes its contribution by addressing a fundamental question: does TQM affect

work attitudes? More specifically, does TQM have an impact on two core attitudes:

teamwork and continuous improvement?

On the surface, the research question seems straightforward and typifies evaluation

driven research. However, and particularly important in the case of TQM, the research

question raises a further question in terms of evaluation: how can the impact of TQM

be assessed given the lack of theoretical propositions underlying TQM which might

serve as a guide to empirical research? In specific terms of this research, how does one

begin to evaluate the impact of TQM on teamwork and continuous improvement

without an explicit set of theoretical underpinnings and propositions to guide such an

evaluation?

In view of this, the starting point for this evaluation is the development of theoretical

models containing hypothesized antecedents of teamwork and continuous

improvement. These models are empirically tested prior to their subsequent use as a

basis for guiding an examination of the impact of TQM on teamwork and continuous

improvement. Therefore, the thesis goes some way towards supplementing the

theoretical basis underlying TQM as well as empirically evaluating the impact of TQM

on attitudes.

The aim of this chapter is to present a critical overview of TQM theory and research. In

doing so, the chapter sets out to accomplish the following. First, to provide a rationale

for investigating TQM prior to examining the nature of TQM. Second, to critically

analyse the theoretical basis of teamwork and continuous improvement. Finally, to

place this research in context by reviewing, in broad terms, the empirical research

2

conducted to date in TQM. In closing, the research questions are presented and the

structure of the thesis is outlined.

1.2 Why investigate TQM?

TQM has been heralded as a "new world order" organizational role model (Boje and

Winsor, 1993). Grant et al. (1994) view TQM as an emerging and distinct management

paradigm that contrasts sharply with the economic model of the firm. In this sense,

TQM is portrayed as a philosophy, "it is the company's raison d'etre" (Grant et al.,

1994, p28). Adopting this interpretation, the authors subsequently argue that

"implementation of TQM therefore provides challenges similar to those involved in the

management of other revolutionary transitions. The management problem with TQM is

analogous to the problems associated with introducing representative democracy into

former autocracies and introducing equal rights into racially segregated societies...."

(Grant et al., 1994, p34).

Other proponents equating TQM with a "new paradigm", a humanistic, systems

approach to management (Brocka & Brocka, 1992), depict customer driven

organizations that are organised around processes and run by teams (Slater, 1991). Ross

(1993) goes further to assert that "no management issue since the Scientific

Management Movement of Frederick Taylor in 1907 has had the impact of the quality

movement" (p2). Therefore, TQM is seen as revolutionary compared to the principles

espoused by the classical management theorists. Clearly, whether TQM can be assigned

new paradigmatic status depends on how TQM is interpreted and how it compares to a

reference paradigm. If one views TQM as a sophisticated modernized repackaging of

Taylor's Scientific Management (Boje and Winsor, 1993), then it would be difficult to

subscribe to the view of TQM as a new paradigm.

3

There is evidence that a new practitioner paradigm is emerging that is characterised by

debureaucratization, downsizing, delayering and decentralization (Kanter, 1992).

However, it would seem that while TQM may not embody all the characteristics of this

emerging paradigm, Hill and Wilkinson (1995) argue that TQM "assists the transition

from bureaucratic to more lissom organizations" (p21). Spencer (1994) argues against

viewing TQM as a new paradigm. In her analysis, she compares TQM to three models

of organizations: mechanistic, systems and cultural. In doing so, the linkages between

TQM and the three models are made explicit. Dean and Bowen (1994) hint at a similar

view asserting that there is considerable overlap between TQM and management theory

and conclude that TQM is a "ubiquitous organizational phenomenon that has been

given little research attention" (p393).

The debate surrounding TQM as a distinct management paradigm partly reflects and

concurrently is affected by a lack of consensus as to what TQM means. As such, it is

quite common to see TQM equated with or described as hazy, ambiguous, (Dean and

Bowen, 1994) and notoriously imprecise (Hill and Wilkinson, 1995). The lack of a

clear conception of what TQM means is not surprising given that the meaning of the

term "quality" is still being debated (Reeves and Bednar, 1994). Despite the plethora of

written material that exists, why is TQM such an elusive concept?

There is no doubt that the contributions of the "founding fathers i " of TQM have paved

the way for the ambiguity and fuzziness surrounding TQM. This is a result of distinct

differences in their approach and prescription. For example, there is a clear divergence

between the proponents on the role of employees; Crosby (1979) assigns a minimal role

while Ishikawa (1985) gives greater emphasis to the contributions of employees in the

There is agreement that this categorization would include Crosby (1979), Deming (1986), Feigenbaum (1983), Ishikawa (1985) and Juran (1989).

4

process of continuous improvement. Hill and Wilkinson (1995) highlight other factors

that have played a role in adding to the chameleon like qualities of TQM. Practitioners

have used TQM as an umbrella term to describe a varied range of practices. Steel and

Jennings (1992) argue that "there are almost as many approaches to TQM as there are

TQM practitioners" (p31). The third factor is the theoretical basis of TQM in statistics

in comparison to the broad social science base of the majority of contributions to

management theory. The consequences of TQM's theoretical grounding (or lack of

such) is pursued later in this chapter.

The chameleon like qualities of TQM are reinforced by Spencer (1994) who argues

that TQM "is not a cut-and-dried reality but an amorphous philosophy that is

continuously enacted by managers, consultants and researchers who make choices

based not only on their understanding of the principles of TQM but also on their own

conceptual frameworks concerning the nature of organizations" (p448). A case in

point is the recent entry of the notion of empowerment into TQM. To my knowledge,

the term is not widely evident in the earlier contributions in TQM 2 while Hill and

Wilkinson (1995) note that it is rarely used in the quality management literature 3

However, in more recent contributions4, it is not only evident but has sparked

considerable debate as to whether, in fact, TQM is a vehicle of empowerment.

2 By early work on TQM, I am referring to the contributions of Crosby (1979, 1986), Juran (1989), Ishikawa (1985), Deming (1986) and Feigenbaum (1983). 3 The authors note that Oakland (1989, p320) refers to empowerment. 4 See for example Grant et al. (1994) and Cruise 0' Brien (1995) for TQM as a source of empowerment and Parker and Slaughter (1993) and Sewell and Wilkinson (1992) for a contrary view.

5

1.3 What is TQM?

Putting aside the opposing views of TQM as a new paradigm, how has it been defined?

At a broad level, it has been defined as a way of managing, a philosophy of

management (Brocka and Brocka, 1992; Dean and Bowen, 1994; Hill, 1991a; Hill and

Wilkinson, 1995; Lawler, 1994) and a systems approach to the practice of management

(Olian and Rynes, 1992; Ross, 1994). There is also agreement that no single theoretical

formulation of TQM exists (Hill, 1991a; Lawler, 1994; Sashkin and Kiser, 1993) nor is

there "a definitive shortlist of practices that are associated with it" (Lawler, 1994,

p68).

Defining TQM as a systemic approach to managing an organization or a philosophy of

management is not particularly informative. Thus, one needs to ask, how has TQM

been characterised? In quite simple terms, TQM can be viewed as comprising a systems

and a cultural component. The first component would include improvement tools,

measurement systems and a quality infrastructure. 5 The latter would include quality

oriented attitudes, values and behaviours. 6 Schein (1985) would argue that these two

components are not and should not be treated as separate but rather the systems

component would be a manifestation of the underlying cultural assumptions.

An organization is said to have to change its culture to facilitate and support TQM

(Brocka and Brocka, 1992; Sashkin and Kiser, 1993; Waldman, 1994). Hill (1991b), for

example, argues that "cultural change is the ultimate objective of quality management"

(p557). What would the notion of a total quality culture consist of ? In other words,

5 This would include improvement teams, cross functional teams, increased customer contact and more autonomous work units. 6 This interpretation of culture is oversimplified and does not reflect a more elaborate interpretation (Schein, 1985) nor does it challenge the notion that culture can be managed (Meyerson and Martin, 1987).

6

what are the cultural values that are important to TQM? Sashkin and Kiser (1993)

outline eight elements of a TQM culture to include, for example, job security, a climate

of fairness, ownership stake for employees, and co-operation as a basis for working

together. Hill (1991b) directs attention to the following elements: "internalization of

quality and continuous improvement as the goal of all activities.... more open

communications....greater involvement.... the creation of high-trust social

relationships....absolute priority of customer satisfaction" (p555). In addition, a move

from individualism to team orientation, from an autocratic management style to

supportive leadership are other cultural values that have been identified as part of a

quality culture (Blackburn and Rosen, 1993).

As can be seen, a number of different cultural dimensions have been put forward as

important to TQM. Some of these dimensions are interrelated (for example, more open

communications and the creation of high-trust relationships) and arguably, some are

more fundamental to TQM than others (for example, continuous improvement

compared to ownership stake for employees). Giving attention to each of the

dimensions portrayed as important to TQM may unduly complicate and overshadow the

essence of TQM. Therefore, the following discussion concentrates on what have been

put forward as the core elements of TQM.

1.3.1 Core elements of TQM

Can TQM be accurately portrayed by a number of key elements? The inherent danger

in reducing TQM to a number of key principles is that the complexity and far reaching

scope of TQM is not accurately reflected. However, by concentrating on a number of

key elements, the essence of TQM can be more easily explicated. Dean and Bowen

7

(1994) assert that TQM is based explicitly or implicitly on three fundamental

principles: customer focus, continuous improvement and teamwork. To a large degree,

Hill and Wilkinson (1995) present a similar interpretation of the underlying key

elements of TQM; these include customer orientation, continuous improvement and

process orientation. The former writers subsumed the principle of process orientation

in continuous improvement. Although teamwork is excluded as a key element by the

latter writers, there is sufficient support for the importance of teamwork elsewhere

(Bowen and Lawler, 1992; Deming, 1986; Hill, 1991b; Wilkinson, 1994). Brocka and

Brocka (1992) argue that "without teamwork, Quality Management is finished before it

can start" (p11). Thus, it may be more accurate to describe TQM as consisting of three

primary but also a number of secondary elements 7 that support and facilitate the

primary elements.

Why are these principles so important to TQM? First and foremost, these principles are

not independent but mutually interdependent. For example, the pursuit of customer

satisfaction (internal or external) may stimulate or provide ideas for improvement

which in turn may require teamwork or co-operative activity between individuals,

groups or departments in order to implement these improvements. Therefore, while

customer satisfaction is the ultimate goal and continuous improvement is the vehicle by

which this is achieved, continuous improvement in itself relies on teamwork.

Continuous improvement requires all organizational members to be motivated to

improve the status quo. In essence this requires a constant questioning and examination

of existing processes in the search for more effective and efficient means of doing

things. As such, commitment to improvement becomes an integral part of an

individual's job; that is, an individual is not only required to carry out his/her specific

These secondary elements may include for example, more open communications, job security, participative management style, collectivist orientation and increased quality awareness.

8

tasks but also to think in terms of how he/she can improve his/her work and that of

his/her work group. In addition to this value or attitude, there needs to be a supporting

system or infrastructure. This could take the form of continuous improvement teams or

at the specific job level, statistical process control methods. Both the value system and

infrastructure are mutually dependent. An individual committed to continuous

improvement needs a vehicle to translate ideas and suggestions into practice and this

may be accomplished through his/her participation in an improvement structure.

Teamwork is interpreted in the widest sense. This includes intragroup teamwork,

interdepartmental teamwork, hierarchical teamwork between management and

employees and finally, interorganizational teamwork between customers and suppliers.

Underlying these forms of teamwork is the notion of co-operative activity. Unless one

has total autonomy over one's work, co-operative activity is necessary to implement

improvements. This is particularly important if an organization's activities are viewed

as an interdependent set of processes. Therefore, a restructuring of work away from the

individual toward a team will highlight interdependence which requires an element of

co-operation.

The notion of interdependence is reinforced with the internal customer-supplier

principle. Employees are expected to view themselves and act accordingly as suppliers

of the next person in the supply chain and customers of the preceding person in the

chain. As Wilkinson (1992) notes, this customer focus is designed to highlight an

individual's contribution to the final customer. Therefore, teamwork and co-operation

are essential at all links in the chain from the initial contact with the external customer

through to the satisfaction of the external customer requirements. "Terms such as

internal customer may reshape members' ideas not only about the relevant

9

organization of work teams, but also about the purpose and importance of their tasks"

(Spencer, 1994, p467).

These three principles of customer focus, continuous improvement and teamwork may

be thought of as comprising a "soft" and "hard" component relating to a cultural and

system dimension. In this sense, soft may be interpreted as values or attitudes while

"hard" relates to the visible mechanisms or structures. For example, commitment to

improvement as an attitude and quality improvement teams as the parallel structure. In

the case of teamwork, a team orientation is paralleled with work restructuring around

the team or ce11. 8 A customer orientation is reinforced by internal customer-supplier

audits and by direct contact with customers.

1.4 A critical analysis of the core elements

The previous section has highlighted what have been portrayed as the important

elements of TQM. This section critically analyses two of the three elements that are the

focus of this research: teamwork and continuous improvement. The rationale for

concentrating on two of the three interrelated elements of TQM is twofold. First, the

intervention (discussed in the subsequent chapter) being evaluated primarily focused on

teamwork and continuous improvement. Second, as outlined at the beginning of this

chapter, the starting point in this evaluation is the development of theoretical models

containing hypothesized antecedents of the key elements. This requires the

measurement of potential antecedents as well as the key elements. In turn, this has

practical implications for the length of the questionnaire. Consequently, an attempt to

measure the three elements as well as their hypothesized antecedents would require a

longer questionnaire which in this case was not deemed feasible. As with any research,

8 This is characterized as a series of self contained mini factories within the factory.

10

a trade-off was made between what was practical in light of the intervention and what

was portrayed as important elements of TQM, in theory.

1.4.1 Theoretical basis

TQM cannot be considered as having an explicit theory; that is, a clear set of theoretical

underpinnings and propositions. This lack of theory is widely recognised and recently,

Dean and Bowen (1994) have argued that theory development is needed to stimulate

empirical research and empirical research should be more productive if there is a

theoretical base to draw on. Sitkin et al. (1994) assert that an inductive strand of

research is warranted whereby from the practices of TQM, implicit theories may be

extracted and made explicit. One of the mechanisms that could assist this is

longitudinal research that examines the effects of TQM interventions. A more

ambitious recommendation for future research is the development of a unified TQM

theory (Steel and Jennings, 1992).

A more critical perspective is adopted by Drummond and Chell (1992) who warn that

TQM could become another discredited initiative if it ignores research findings from

disciplines such as Organizational Behaviour (OB). Kerfoot and Knights (1995) go

further to argue that TQM is fundamentally flawed by the inherent contradictions that

arise in the implementation of TQM initiatives. "These contradictions may be seen to

derive from the crude and mechanistic 'engineering-like' model of organizations and

the equally simplistic understandings of human behaviour that inform TQM" (Kerfoot

and Knights, 1995, p220).

There is an implicit assumption that a TQM initiative will be interpreted similarly by

employees and also, that they will respond in a similar manner. Thus, there is a lack of

11

attention paid to individual differences and the role these differences play in how

employees perceive and respond to TQM interventions (Kerfoot and Knights, 1995).

This type of assumption reflects and is indicative of the embryonic stage of theoretical

development of TQM. What affects teamwork and commitment to improvement?

There is no explicit set of theoretical propositions within the TQM framework that help

explain why some individuals may have a stronger team orientation than others or why

some individuals may be more committed to improvement than others. In other words,

what are the antecedents of teamwork and commitment to improvement? More

fundamentally, what does teamwork and commitment to improvement mean? These

questions largely remain unanswered in the TQM literature.

1.4.2 Teamwork

If teamwork is applied to the work group, what does this mean? It is widely used in the

TQM literature and its meaning is assumed to be self evident. However, teamwork can

represent a form of work design based on a group or it may be interpreted as an

individual's orientation toward the group and how individuals within the group interact.

Therefore, the two interpretations reflect a 'hard' and 'soft' dimension; that is, the

former emphasizes the structure while the latter emphasizes teamwork as a value or

attitude. In terms of labelling, the latter dimension may be better reflected under the

guise of team orientation. Work may be structured around a team thus highlighting the

interdependence of tasks and the importance of co-operative activity. However, is work

restructuring enough to create a 'team spirit' within the group?

It would seem that if there is any theoretical basis to team orientation in TQM, it is

likely that an assumption is held that work redesign around the team will lead to team

orientation. However, Wickens (1995) argues that teamwork and work groups are

12

distinctly different. "Teamworking is a culture; work groups are about structures"

(p119). Furthermore, creating a work group structure does not automatically mean that

teamwork will result. Teamwork in this sense is analogous to team orientation. There

is no explicit theory underlying team orientation in TQM. This raises a number of

issues. First, what are the antecedents of team orientation? Contrary to the broad

assumption underlying TQM that individuals will adopt values and attitudes in a similar

manner (Kerfoot and Knights, 1995), what differentiates individuals in terms of their

team orientation? Second, does a TQM intervention lead to a significant improvement

in team orientation? If so, how and why does this occur?

These questions are left unanswered in the TQM literature and furthermore, there is a

scarcity of empirical research that systematically addresses these issues. Cruise 0'

Brien (1995) found that employees were more willing to exert effort on behalf of their

team rather than the organization. What the study does not address is the factors that

affect team orientation.

Research findings from other areas would suggest that there are antecedents of team

orientation. These are dealt with more fully in chapter 4 but for the purpose here, two

factors serve to illustrate the existence of possible antecedents. First, interpersonal trust

has received scant attention in the TQM literature (an exception would include Hill,

1991 b) and this has been found to have an effect on team cohesiveness and co-

operation (Golembiewski and McConkie, 1975). The greater an individual's trust in

their colleagues, the more likely that he/she will hold a stronger team orientation. In a

similar vein, the role of the supervisor may influence or shape an individual's

orientation toward the work group. This underlies Human Relations theory which

proposes that leaders using participation can develop work group members into a

working team with high loyalty (Likert, 1961). Thus, in theory, it is possible to identify

13

factors that may affect an individual's attitude toward the team. From this, one would

not necessarily expect all individuals to have a similar team orientation as a result of a

TQM intervention.

How would a TQM intervention affect team orientation? If the above holds true; that

is, trust and the participative style of the supervisor affects team orientation, then if the

intervention affects trust and participative style, there is one possible linkage (in theory)

between TQM and team orientation. While the TQM literature is quite explicit in its

prescription of change for supervisory behaviour, it does not link this change to specific

changes in team orientation of subordinates.

Therefore, in theory, there are possible antecedents of team orientation and it is possible

that a TQM intervention may affect team orientation through these antecedents.

Consequently, while the TQM literature does not provide an explanation of how team

orientation may be affected, it would seem that by borrowing from previous research

findings and theoretical insights, this gap may be filled.

1.4.3 Continuous improvement

A critical assessment of commitment to improvement is further complicated by

different contributors emphasizing different concepts. Specifically, some writings

emphasize intrinsic motivation and others continuous improvement.

In theory, employee commitment to improvement represents a change in their work;

they are not only required to do their job in the most efficient and effective manner but

also think of ways in which their work and that of their work area may be improved.

Thus, placing the onus on employees for improvements adds an additional component

14

to their work. Therefore, if one equates eliciting commitment to improvement with

redesigning work, will all employees respond in a similar manner by integrating the

search for improvements into their daily tasks?

TQM requires the participation of all organizational members in the pursuance of

continuous improvement. Overall, there is a consensus that total involvement is a

requirement (Ishikawa, 1985; Lawler, 1994; Oakland, 1989) translating into an

organizationwide commitment to continuous improvement. Financial incentives to

induce employee participation in TQM or commitment to continuous improvement are

completely eliminated by many of the quality proponents (Crosby, 1979; Deming,

1986; Oakland, 1989). This type of inducement "does not form part of a TQM culture,

and would defeat many of the objectives" (Oakland, 1989, p303). Hill and Wilkinson

(1995) assert that "the belief of the quality gurus is that people need to buy into TQM

without coercion, because they have internalized a commitment to quality management

and voluntarily pursue the appropriate principles and practices" (p14). In particular,

TQM proponents are highly critical of conventional performance related payment

systems that focus on the individual. First, TQM proponents argue that individual

performance related pay systems focus an individual's attention and effort on obtaining

the rewards and consequently, individuals may attempt to set less challenging goals.

Kohn (1993) argues that this type of reward elicits temporary compliance, it does not

create an enduring commitment to any value or action. Second, individual self interest

may be pursued at the expense of teamwork and continuous improvement.

Kelman (1958) argues that high levels of commitment are dependent upon the

individual internalizing the organization's goals and values. In the context of TQM,

this involves the internalization of values such as teamwork and continuous

improvement. The use of reward systems to elicit desired behaviours results in

15

compliance which negates the desired result of commitment (Drummond and Chell,

1992). Therefore, the use of coercion and sanctions goes against the commitment that

TQM seeks to achieve.

What happens in practice? Do employees internalize the values of continuous

improvement? Do they voluntarily and willingly participate in TQM? Recent case

study evidence (McArdle et al., 1995) questions the willingness of employees to

voluntarily participate in Quality Circles (QCs). While membership in QCs was

voluntary, overt and covert encouragement was given to employees to participate. At

this particular plant, wage increases for employees were based on performance

comprising a quantitative and qualitative element. In terms of the latter component,

this was assessed on behavioural characteristics. "Being a member of a quality circle

and taking an active part in the quality process is seen as desirable behaviour, which

will be rewarded accordingly" (McArdle et al., 1995, p162). Moreso, there is a hint

that the idea of voluntary participation is no longer appropriate. Soin (1992), an

advocate of total quality, argues that in today's competitive environment,

[employee] "participation is no longer voluntary-it is essential for success" (p 211).

This raises a question of how employee commitment to continuous improvement is

achieved. What factors affect an individual's commitment to continuous improvement?

If commitment to improvement involves changing the nature of jobs, research on work

redesign (Hackman and Oldham, 1980) may serve as a guideline. Not all individuals

will respond similarly to work redesign. In particular, an individual's higher order

need strength has been found to moderate the relationship between work redesign and

its associated outcomes. Consequently, one could hypothesize that an individual's

higher order need strength may influence how individuals respond in terms of their

commitment to improvement. From this, individuals with a greater need for

16

achievement and satisfaction through work may be more likely to be committed to

improvement.

Along similar lines, an individual's commitment to the organization may have a

positive effect on commitment to improvement. Recent empirical evidence suggests

that organizational commitment has a significant positive impact on quality

consciousness (Peccei and Wood, 1994; Wood and Peccei, 1995). Therefore, if

commitment to continuous improvement is perceived to be a significant value or goal in

the organization, one could hypothesize that the more committed an individual is to the

organization, the more he/she will be committed to continuous improvement.

Several issues need to be addressed in relation to commitment to improvement. First,

what does the concept mean? How can it be operationalized? Second, what are the

antecedents of commitment to improvement? Does TQM significantly affect

commitment to improvement? Is commitment to improvement the same as intrinsic

motivation?

Two opposing views are briefly presented concerning the nature of intrinsic motivation

and commitment to improvement. First, one could argue that the two concepts in

essence are tapping an individual's desire to do the best possible job they can. If this

view is accurate, the two concepts may be used interchangeably as seems to be the case

in the TQM literature with some contributors emphasizing continuous improvement

and others intrinsic motivation. However, a contrary view is hinted at by Lawler (1994)

who argues, in the context of employee involvement, that, employees have the

responsibility of calling attention to problems that prevent them from doing a quality

job thus preventing them from being intrinsically motivated; they also have the

responsibility of accepting a continuous improvement culture. The conceptual

17

similarity of intrinsic motivation and commitment to improvement is discussed and

empirically tested in this thesis.

1.4.4 Change

The previous discussion highlighted the lack of explicit theories underlying the key

elements of TQM. This section builds on this by looking at how an organization

achieves team orientation and commitment to improvement. In other words, what is the

approach to change underlying TQM? If one accepts that these attitudes or values are

consistent with a total quality culture, how is culture affected?

Putting aside the debate on whether or not culture can be managed (Ogbonna, 1992-

1993, Williams et al., 1993), given that culture is central to TQM, what are prescribed

as the mechanisms to achieve a "total quality culture?" How does change occur within

organizations? Hill (1991b) argues that the quality proponents rely on training,

education and leadership as the means to affect culture. This is viewed as adequate to

create the appropriate culture. Hill (1991b) argues that attention should be directed to

the work of organizational behaviour academics in the area of culture. In specific

terms, Schein (1985) presents primary and secondary levers to change culture.

Consistent with the quality proponents, leadership and education are included.

However, Schein (1985) argues for the deployment of organizational rewards and

punishments as a primary mechanism for affecting culture. The use of rewards and

sanctions goes against the philosophy of TQM. However, they may be necessary to

show that TQM is taken seriously and to induce behaviour change in the initial

instance. The rationale provided by the quality proponents is that rewards and

sanctions produce compliance and not commitment.

18

In terms of organizational change, TQM relies primarily on nollnative-reeducative and

empirical-rational strategies (Chin and Benne, 1976) to affect commitment to core

values and attitudes of TQM. Implicit is the assumption that a change in attitudes and

values will facilitate a change in behaviour. However, serious doubt has been raised

about the effectiveness of this approach (Guest, 1984). The causal path between

attitudes and behaviour may be more complicated. Rather than a unidirectional link, it

may be a relationship of mutual influence. As such, changes in behaviour may

facilitate attitudinal change.

Wilkinson (1994) argues that an organization's existing culture may inhibit the

implementation of TQM. This view is also voiced by Snape et al. (1995) in that an

existing culture may be a source of resistance and act as a barrier to the successful

implementation of TQM. An existing organizational culture may not be the sole reason

for ineffective change efforts. One factor that may contribute to the ineffectiveness of

change may be a lack of readiness for change. This notion of readiness is similar to

Lewin's (1951) concept of unfreezing and reflects organizational members' beliefs and

attitudes concerning the need for change and the organization's capacity to implement

the changes (Armenakis et al., 1993). The importance of readiness is supported by

Schein's (1979) argument, "...the reason so many change efforts run into resistance or

outright failure is usually directly traceable to their not providing for an effective

unfreezing process before attempting a change induction" (p 144). The role of a need

for change is underplayed in the TQM literature and this may help explain why TQM

initiatives may not succeed in affecting change.

In addition to acting as a source of resistance, an organization's culture may influence

how an organization interprets TQM and consequently the practices that are adopted.

In simple terms, a bureaucratic oriented organization may be more inclined to interpret

19

and view TQM in a mechanistic way. Particular emphasis may be given to

implementing (as a starting point) TQM as a set of techniques to improve

organizational efficiency.

Champions of the TQM approach adopt an evangelical stance in terms of employees

embracing the values of TQM with blinding faith. Based on this view, employees will

view TQM in a positive light and consequently adopt the appropriate attitudes. The

rationale is that TQM brings benefits in terms of: a repudiation of scientific

management, employee empowerment, involvement and participation. Do employees

see TQM as beneficial? Some employees may interpret TQM as increasing

responsibility, stress and work intensification without a parallel increase in pay. Boje

and Winsor (1993) argue that TQM is a sophisticated repackaging of Scientific

Management. Employee suggestions for improvement are interpreted as a mechanism

in which individuals "taylorise" their own work. In addition, rather than relying on

hierarchical control, the notion of teamwork allows control to be exercised by peers

whereby group deviants are compelled to conform. Therefore, the key argument is that

in TQM, instead of relying on management to "taylorise" work, employees are

empowered to self-taylorise their own work.

Kerfoot and Knights (1995) assert that the quality literature neglects to consider that

TQM interventions may be interpreted differentially by employees. Thus, individuals

may differ in terms of how they perceive the intervention. Some individuals may see it

as more beneficial than others. Empirical research on organizational commitment

suggests that affective attachment to the organization would positively colour an

individual's assessment of organizational actions and activities (Eisenberger et al.,

1990). This would indicate that the commitment of individuals to the organization are

likely to predispose them to viewing the intervention in a more or less positive manner.

20

How individuals perceive a TQM intervention may affect the degree to which they

adopt attitudes and behaviour consistent with TQM. Recent evidence suggests that it is

not employee participation per se that has a significant effect on quality consciousness

but rather employees' experience of their participation (Peccei and Wood, 1994; Wood

and Peccei, 1995). If this holds true, attention may need to be diverted away from

eliciting employee participation using whatever means to ensuring that the intervention

is seen as providing some benefit to employees.

A review of the theory on TQM raises more questions that it provides explanations.

This lack of theory has implications for evaluating TQM interventions; there is very

little by way of guidelines as to how and why such interventions affect the key elements

of TQM Consequently, any evaluation of a TQM intervention would have to begin

with a theoretical development of models that help explain core phenomena such as

teamwork and continuous improvement.

1.5 Empirical research

Research in TQM thus far can be broadly classified as organizational level surveys and

qualitative case studies. The driving force behind the former type of research is an

assessment of the impact of TQM interventions. The most prominent large scale

survey into TQM practices is the series of surveys conducted by the Center for

Effective Organizations at the University of Southern California9 . The following

9 Three surveys (1987, 1990, 1993) were administered to 1000 top Fortune organizations. The initial focus of the research was approaches to employee involvement which was subsequently modified to include TQM practices reflecting its growing importance.

21

discussion focuses on the results of the latest survey (Mohrman et al., 1995) which

primarily examined TQM practices and their effect.

The results of the survey indicate that a high percentage of organizations (73%) are

implementing TQM, covering on average 50% of employees and more prevalent in

manufacturing than service organizations. Overall, TQM experience has been positive

and beneficial in terms of productivity, competitiveness, profitability in addition to

employee satisfaction. Thus, it would seem that all stakeholders are benefiting from the

adoption of TQM. While the study represents the most extensive examination of TQM

practices in the U.S, the findings are subject to the limitations of cross sectional

research. In addition, there are risks in relying on a single perception or viewpoint as

representing the organization's activities and adoption of TQM practices. Nonetheless,

the research provides an overall picture of TQM adoption in the U.S.

An equally extensive survey has not been conducted in the UK. However, a number of

large studies have examined the impact of TQM (for a useful summary, see Hill and

Wilkinson, 1995). In contrast to the positive portrayal of the U.S experience, the

evidence in the UK suggests that TQM has fallen short in its delivery of benefits. This

may be due to the paucity of organizations with "full blown" TQM and the extensive

adoption of TQM in a less than total orientation (Wilkinson et al., 1992, 1993; Cruise

0' Brien and Voss, 1992). Kearney (1992) found that 80% of organizations surveyed

reported that TQM initiatives failed to produce any tangible benefits. Similarly,

Wilkinson et al. (1993) found that only 8% of a large sample of managers in the UK

claimed that quality management was successful. How can these rather negative

findings be interpreted? Are the results due to the piecemeal adoption of certain TQM

practices that lack coherence or integration? Is it that adequate time has not lapsed in

order that TQM can fully develop? Is it the search for quick results stimulated by

22

promises from consultants (Hill and Wilkinson, 1995) that when they are not achieved

result in disillusionment with TQM?

Overall, these organizational level surveys are useful in providing an overall picture in

relation to the adoption of TQM practices and their impact. This type of research needs

to be complemented with case studies that provide greater detail and are more

analytical in orientation. The empirical findings of a number of case studies are now

briefly discussed.

The initial focus of case studies was on TQM in practice; how organizations

implemented TQM (Hill, 1991b; Wilkinson et al., 1990; 1991). Subsequent case

studies examined specific issues or aspects of TQM. McArdle et al. (1995) explore

issues such as empowerment and involvement in the context of TQM. The results

suggest that empowerment did not occur in terms of extending employees' rights, rather

work intensification in conjunction with greater monitoring and control of work

occurred. Webb (1995) concentrates on the role of management in TQM. Her case

studies suggest that TQM has radical implications for managerial and technical roles.

More recently, the relationship between HRM and TQM has been receiving increased

attention (Snape et al., 1995; Wilkinson, 1994) and Simmons et al. (1995) provide case

study evidence exploring the successes and problems encountered in one organization

in employing strategic HRM to pursue TQM.

The case studies conducted to date have primarily adopted a qualitative approach and

have a tendency toward ex post facto analysis of the effects resulting from the

implementation of TQM. More importantly, there is a noticeable absence of

longitudinal studies aimed at systematically evaluating TQM interventions. However,

exceptions do exist. As previously mentioned, one study investigating the impact of a

23

TQM intervention on quality consciousness found that it was not employee

participation per se that had an impact but rather how employees assessed and judged

the intervention ( Peccei and Wood, 1994; Wood and Peccei, 1995). Guest and Peccei

(1994) found that a quality improvement intervention did not have a significant effect

on organizational commitment.

1.6 Research objectives

This chapter has reviewed the literature on TQM with particular emphasis on what have

been heralded as some of the important elements of TQM. Two issues emerged from

this review. First, there is little theory underpinning the elements of TQM. Second,

there has been little systematic investigation into the effects of TQM on some of these

elements.

In view of the relatively early stage of theoretical development and a lag in empirical

investigation in TQM, potential research questions abound. The core research question

of this thesis is: does TQM affect teamwork and continuous improvement?

As mentioned in the opening of this chapter, the starting point in this evaluation is the

theoretical development and subsequent testing of models containing hypothesized

antecedents of teamwork and commitment to improvement. The examination of the

antecedents of teamwork and continuous improvement as the first stage in this

evaluation is important for both theoretical and methodological reasons. First, the

development and testing of 'antecedent' models goes some way towards developing

explicit theories and supplementing the theoretical basis underlying TQM. Second,

from a methodological stance, it is important to control for other effects that may have

24

an impact on teamwork and continuous improvement. In particular, the effects of the

hypothesized antecedents need to be controlled for in order to rigorously determine the

direct effects of TQM.

Finally, we do not know, from existing research, whether TQM directly affects

teamwork and continuous improvement or whether it has an indirect effect (in addition

to or instead of a direct effect) on the two elements through affecting their antecedents.

Therefore, the antecedents provide a way to capture some of the indirect effects of

TQM on teamwork and continuous improvement. This is particularly important given

that few evaluations have been conducted and consequently, there is a lack of

knowledge as to how TQM affects the presumed outcomes.

TQM can be thought of as having two dimensions: employee participation in and

assessment of TQM. The first dimension captures a behavioural component of TQM

while the latter taps an affective component; how employees judge and assess TQM.

Therefore, the core research question examines the effect, if any, of individuals'

participation in and assessment of TQM on teamwork and continuous improvement.

In order to provide a greater understanding of the potential impact of TQM, it is

necessary to ask what factors, if any, may affect an individual's participation in and

assessment of TQM. While this question is not evaluation driven per se, it is important

in light of the organizationwide emphasis of TQM and its reliance on voluntary

participation at the employee level.

25

1.7 Structure of the thesis

This chapter presented a critical overview of the TQM literature and outlined the

research questions (these are discussed in greater detail in chapter 3). The subsequent

two chapters deal with; the context in which this research was undertaken, the content

of the TQM intervention and the process of implementation (chapter 2); the approach

taken to the gathering of data and the logic underlying this evaluation (chapter 3).

Together, these initial chapters set the scene and provide the basis for presenting the

results.

Chapter 4 examines the antecedents of team orientation prior to investigating the

impact of the intervention on team orientation in chapter 5. As a precursor to

examining the impact of the intervention on commitment to improvement (chapter 7),

chapter 6 examines the conceptual similarity of commitment to improvement and

intrinsic motivation. Chapters 8 and 9 investigate the predictors, if any, of employee

participation in and assessment of the TQM intervention.

The common element of the results chapters thus far is that they focus exclusively on

employees. Consequently, chapter 10 examines change, if any, in supervisory

behaviour and the predictors of two key dimensions of behaviour; participative style

and commitment to quality. Chapter 11 integrates the findings by making an overall

assessment of the intervention using a number of different criteria. The subsequent and

final chapter draws the conclusions and suggests directions for future research.

26

Chapter 2: The Context

2.1 Introduction

Having raised the research questions in the previous chapter, this chapter presents a

descriptive account of the organizational context, the content of the TQM intervention

and the process of implementation. The subsequent chapter builds on this by

presenting a detailed overview of the research methodology employed in this study.

Together, these two chapters bridge the research questions posed and the empirical

findings.

Providing an account of the organizational context and the TQM intervention is

important for a number of reasons. First, and foremost, as a consequence of the lack of

conceptual clarity of TQM, TQM interventions may take on a variety of forms that

could be expected to lead to different outcomes. Second, the content of the intervention

and the context in which it occurs may assist in the interpretation of the results and also

have policy implications for organizations implementing TQM interventions. Finally,

the content of the intervention may be a key explanatory factor in comparing the

outcomes of TQM interventions in different organizations.

The initial objective of this research was to compare two sites within the same

organization that differed in terms of their progress in implementing TQM.

Consequently, a site was selected that was about to embark upon a TQM intervention

and as a contrast, a site was chosen that had progressed considerably in terms of

implementing TQM. While the two sites were subject to the same organizationwide

reorganization of manufacturing operations and changes in employee practices and

policies (this is discussed in detail later), there was an interesting difference in terms of

27

the emphasis given to the so called 'hard' and 'soft' dimensions of TQM. The site

embarking upon the TQM intervention emhasized the cultural dimension while the

other site, although espousing the importance of attitudes and values, placed greater

emphasis on the systems associated with TQM. Due to unforeseen circumstances

which are briefly described in the subsequent chapter, it was not possible to include the

second site in this manner in this study.

This chapter begins by describing the overall organization and the site implementing a

TQM intervention. This is followed by a brief historical portrayal of the changes that

occurred prior to the TQM intervention. Subsequently, the rationale for embarking

upon the intervention is presented. Following from this, the content of the intervention

and its process of implementation (for the duration of this study) is reviewed.

2.2 The Organization

The organization is a UK based multinational supplier of engineering and electrical

components. Within the UK automotive industry, the organization dominated the

vehicle components market in the 1960s. Its dominant position remained during the

1970s and 1980s although the absolute size of the market shrank catastrophically. The

most profitable aspects of the business were the overseas operations while the

automotive components side incurred a loss for the first time in 1981 in over a century

of trading. In response to this, operations were streamlined, sites were closed and

amalgamated and the labour force was reduced. The reduction in the work force was an

immediate response to the falling demand but also provided the foundation for

subsequent changes geared towards improving labour productivity.

28

Thus, the operating sites within the automotive components side of the organization, as

a precursor to a major reorganization of production, adopted new technology, a tough

`take it or leave it' approach to pay negotiations, a shedding of the work force and the

introduction of minor changes to work practices. Together these changes provided the

basis for the introduction of a new 'Japanese style' manufacturing method as part of the

Survival Plan of the mid 1980s. The story is now picked up by focusing on the site in

this study.

2.2.1 The Site

The site was one of 34 production units of the company in the UK employing 1,100 in

the early 1980s. 1 In a personal message to all employees in 1985, the General Manager

of the site stated "despite the strenuous actions on part of all employees over the past

few years we are still trading at a loss and, unless this situation is remedied, we must

eventually go out of business" (internal documentation). The major reason for the

continuing losses of the site was a dramatic decline in the market for heavy duty

electrical components against a background of increased competition. In the previous

five years, the UK market for heavy duty electrical equipment fell by 48%. This was a

direct result of a reduction in the UK bus and truck production. While the size of the

market shrank, the site was able to increase its market share at the expense of

competitors.

In order to reverse the operating losses, fundamental changes were needed. Labour

costs were 50% of total outgoings so a reduction in the labour force was on the cards. 2

In addition, the remaining objectives included the production of the highest quality and

2 In the previous decade, the work force had been reduced from 2,300 to 1,100. The number of production units in the UK is currently approximately 27.

29

most reliable equipment, and a reduction in the response time to fulfilling customer

requirements.

2.2.2 Major changes at the site

The new way forward for the site was the introduction of a new manufacturing

technology. The site was not the first within the overall organization to introduce

modular manufacturing systems. It was adopted as an organizationwide move toward

more flexible manufacturing systems. Module production (originating from the

`kanban' system in Japan) is a method akin to lust-in-time' production where stocks

and work in progress are aligned to fit the production schedule. The system has been

described as mini factories within the factory (Turnbull, 1986).

After a pilot scheme, the modular system was introduced throughout the site in 1988.

Each module could be described as a mini business producing a single product or a

single group of products. All the manufacturing processes associated with that product

were placed within the module and also, other activities which were necessary to the

module were designed to give maximum support. This involved the transfer of

activities traditionally carried out by indirect support departments to the module itself

For example, rather than having a centralised buying or purchasing department, this

function was devolved and integrated into the module as one of its activities. However,

activities such as personnel and management services remained centralised. Depending

on the number of products manufactured within the module, the module may have been

divided into a number of cells corresponding to each product or component. This cell

based structure, otherwise known as teamworking, became the building block of the

new system and represented a radical departure from previous production methods.

30

A key feature of the new system was that personnel within the cell or module worked as

a team. In some cases, this required a wider range of skills from employees. All

employees were required to be willing to do whatever they could (taking into account

their training and capabilities) to keep their cell or module operating. The aim was to

have as much flexibility within the cell or module as possible. In practice (among other

things), craftsmen and operators were trained to set up and operate respectively as many

different machines as the cell or module required to manufacture their component or

product. One of the major consequences of this new system was a significant reduction

in the number employed. Inherent in the operation of the new system was flexibility

not only in technology but also in personnel. What emerged was the disappearance of

demarcation, the breeding of 'super craftsmen' and the upgrading of operator skills.

In terms of organizational structure, there was a flattening of the hierarchy. The

management structure was reduced from a seven tier to a five tier structure.

Classification of direct production employees was reduced from seventeen to two.

Thus, after the manufacturing reorganization, production employees were either

craftsmen or operators. Through a combination of redundancies and retraining for new

jobs, the number employed at the site was reduced by 50% (from 1,100 to

approximately 580) by the late 1980s.

In parallel to the manufacturing reorganization, changes in terms and conditions of

employment were introduced. Specifically, this involved the creation of single status

employment with no difference in conditions and benefits between staff and works

employees. Methods of payment and pension scheme were harmonised. Pay grades for

the shop floor were reduced from 59 to 8 and a simplified payment for skills was

introduced.

31

By 1990, the site had transfoirned the loss making situation to one of profitability. The

product range was streamlined from 121 product types to 66. Responses to customer

needs were reduced from on average three months to 1-6 weeks. Overall, the results of

the manufacturing redesign meant that the site could match its competition on quality

and delivery. From this point, the main strategic objective was to focus on a narrower

product range and to target European vehicle manufacturers for growth. Further

changes were introduced to include continuing education and training schemes and an

open learning centre. This reinforced the notion of continuous improvement to ensure

the survival of the site. In conjunction with healthcare, these changes signalled a

willingness to invest in employees in order that they may reach their potential which

would benefit the site. A comprehensive communications programme was launched

which served (among other things) to bring what was happening in the market place to

the door of the manufacturing cells and modules reinforcing the business focus of the

new manufacturing system.

Compelled to build on the momentum of previous changes, the site in early 1991

launched Continuous Improvement Groups (similar to the widely known Quality

Circles) as a means of continuing the improvements already achieved. From

discussions with management and employees, the general view was that this initiative

was doomed to failure from the start for a number of reasons. The main managerial

objective in introducing Continuous Improvement Groups was to provide a mechanism

for employee contribution to efficiency and quality objectives. These groups never

gained a strong foothold at the site. From management's view, participation was

voluntary and there was a lack of employee willingness to participate. Some employees

felt that they had contributed to the site by accepting the reorganization of production

and were unwilling to voluntarily participate in further change. Probably, the primary

reason for the failure of these groups was a lack of visible support from managers and

32

supervisors. Aside from the perceived lack of importance attached to these groups,

they were interpreted as an add on activity rather than integrated into the normal

activities of the site. At this stage, the site was now making a profit which

consequently did not help in creating at supervisory or employee levels a perceived

need to change. Overall, the take up of these groups was sporadic with some groups

disbanding while others started.

By the spring of 1992, twenty or so employees were participating in these groups which

were slowly fizzling out. In fact, some employees were unsure if their Continuous

Improvement Group was still in operation. Overall, the experience and failure of these

groups prompted a more serious organizationwide endeavour into TQM. Against the

background of the major manufacturing reorganization and the failure of Continuous

Improvement Groups, the site embarked upon a TQM intervention under the title of

`Working Together to Win' (WTTW). The remainder of this chapter concentrates on

the content and process of this intervention. However, as a prelude, the thinking and

intentions behind the intervention are discussed.

2.3 Rationale behind WTTW

Overall, the intervention was seen by the executive management team at the site as a

natural progression of previous changes. Furthermore, while the prior changes

primarily focused on the hard visible reorganization of manufacturing, it was felt that

the culture of the site lagged behind in terms of progress. Consequently, in order to

guarantee the survival of the site and to ensure its continued profitability, the key was

to change the culture of the site. While this was not the objective of the Continuous

Improvement Groups, the lessons learned from this experience were twofold: the

change needed to occur throughout the site and it must start from the top. This type of

33

transition from Continuous Improvement Groups or a similar grassroots improvement

structure to TQM is not uncommon. Hill (1991b) found that in the initial instance,

some organizations experimented with Quality Circles prior to implementing TQM.

Even though the foray at the site with Continuous Improvement Groups was far from

successful, it signalled the potential value of employee involvement in continuous

improvement and thus the way forward.

The intervention was perceived as a natural progression of the previous changes in

work methods, systems and structure. The view of TQM as a philosophy or culture was

a key factor in the design of the intervention. The objective was continuous

improvement and this was to be achieved by the participative involvement of everyone.

Therefore, continuous improvement and continuous change were seen as providing the

key to the future of the site. This relied on people changing their attitudes toward

continuous improvement; viewing continuous improvement as an integral part of their

job.

By late spring of 1992, the executive team had enlisted the help of a Total Quality

expert from within the organization. This individual had considerable knowledge of the

site and viewed TQM more in tern's of attitudes and values rather than systems and

techniques. The role assigned to this individual was to act as a bridge between the

executive team and the group of outside consultants that would design and implement a

`soft' TQM intervention. Shortly thereafter, while the forthcoming TQM intervention

was in the early planning stages, the baseline questionnaire was completed by the

participants in this study. With the exception of the executive team, no-one at the site

was aware or had knowledge of the forthcoming intervention.

34

Considerable time and energy was given to the design of the intervention; the content

of the training programme and its subsequent implementation. The overriding

objective of the intervention was cultural change aimed at continuous improvement

through the vehicle of 'participative involvement'.

2.4 The TQM intervention

What constitutes a TQM intervention? How does the intervention at this site compare

to other interventions? This raises a broader issue of what changes are part of and what

changes remain outside TQM. Hill and Wilkinson (1995) argue that among other

contingent factors, TQM may manifest itself differently depending on the stage of

quality development. Reflecting different views of TQM, what may be considered part

of a TQM intervention in one organization may be viewed as the opposite in another

organization. To illustrate this point, McArdle et al. (1995) report from their case study

that while TQM was introduced in 1986, harmonization of benefits and conditions did

not occur until 1988 and appears to be part of the TQM philosophy. In this study,

harmonization occurred in parallel to the manufacturing reorganization which paved the

way for TQM. The first venture into "TQM" at this site (as the case with a number of

other organizations, see Hill, 1991b; Rees, 1995) was through Continuous Improvement

Groups.

While the restructuring of work along cellular lines at the site is consistent with TQM,

it did not occur as part of TQM but rather as part of an organizationwide move to

restructure the manufacturing basis of its operations. McArdle et al. (1995) found that

"at the operator level the main changes which were introduced into the operation were

as much a part of the changes in the manufacturing systems as to do with the TQM

system" (p162). What this illustrates is the difficulty in disentangling the effects of

35

TQM and the ambiguity surrounding what is part of a TQM package. Most of the

empirical work on TQM takes the form of retrospective case studies spanning years

rather than real time monitoring of the implementation of TQM. Consequently, this has

implications for analysing what is introduced as part of TQM and what is introduced as

part of other changes.

Jenkins et al. (1995) in their case study of TQM in Royal Mail noted that a TQM

programme was embarked upon in 1988 while a business reorganization which

included delayering and resulted in voluntary redundancies occurred in 1992. Thus, the

softer dimension of creating a climate for continuous improvement was a precursor to

the hard structural reorganization. In essence, the sequence of changes occurred in the

opposite direction to the changes in this study.

Returning to the issue of what a TQM intervention involves, it may be beneficial to

classify types of interventions akin to the classification of HRM strategies. Wilkinson

(1992) interprets the 'hard' side of TQM to comprise the tools and techniques of TQM

while the 'soft' side as creating customer awareness through educative means. This

type of broad categorization could be applied to TQM interventions so that the

emphasis could be more clearly marked. Thus, the 'hard' dimension may include the

production tools and techniques of TQM, structural reorganization and quality

infrastructure necessary for continuous improvement. In contrast, the 'soft' component

of TQM may be more concerned with instilling the values and attitudes of a total

quality culture. There is a danger in attempting to classify interventions as one or the

other thus viewing the two dimensions as distinct. In reality, many interventions

include elements of both and as the previous chapter suggested the principles of TQM

may have a 'hard' and 'soft' component relating to a system and cultural dimension.

36

Although crude, the above classification of TQM interventions, helps to highlight

where the emphasis of TQM interventions lie. In this study, the emphasis was clearly

on changing the attitudes and values and on the creation of a total quality culture. As

such, the objective of the intervention was similar to Xerox's leadership through

Quality process "aimed at fundamentally changing the way Xerox people work and

manage so they can continuously improve the way they meet the requirements of their

customers" (Ross, 1994, p53).

In attempting to create an involvement culture, the training and education covered such

issues as Theory X and Y, leadership styles and empowerment. Specifically, the focus

was on changing managerial assumptions which in turn would lead to behaviour change

thus influencing the behaviour of those in subordinate positions. Therefore, the

primary vehicle for achieving culture change at the site was through changing the

attitudes and behaviour of those in supervisory positions. This thinking is consistent

with that in the TQM literature whereby managers are the initial target group for change

and change at this level is a necessary prerequisite for successful TQM efforts (Hill,

1991b).

A second focus of the education and training was on leading and managing groups.

This entailed team building exercises, techniques for effective team meetings and

problem solving techniques. An extension of this was work improvement through

leaders facilitating employee involvement in the improvement process. Continuous

improvement is the job of all employees who can contribute to small scale

improvements known as 'kaizen'. Naturally occurring work groups can also contribute

to `kaizen' but in addition to cross functional teams may make breakthroughs in terms

of major improvements.

37

The traditional TQM tools and techniques included the Deming Cycle, Quality Grid,

Cost of Quality, Customer- Supplier process, Brainstorming, Fishbone Diagrams and

Force Field Analysis. These tools were included as guides to achieving work

improvements. Finally, structuring expectations, measuring performance and

participative objective setting were included. The training and education programme

was aimed at kick starting the culture change at the site.

The education and training programme was perceived to be a sufficient stimulus and

inducement for culture change at the site. In theory, supervisors who completed this

programme would reproduce the training for their subordinates. Following from this, a

WTTW team would be set up in that particular work area which would give employees

a vehicle for greater say in what happens in their area, provide a mechanism for work

improvements stimulated by the internal customer-supplier audit completed

periodically with the internal customers of the work area. Furthermore, it would

provide a forum whereby problems individuals faced in doing their work could be

eliminated and suggestions for improvements could be implemented.

Therefore, it was assumed that as a result of the training programme, a series of

subsequent changes would occur throughout the site. The programme would provide a

common objective between managers and employees, departments and work areas and

a common 'language' thus providing the foundation for continuous improvement. In

addition, an underlying assumption (also visible in the TQM literature) was that

education, training and leadership were sufficiently strong tools to affect change.

38

2.5 The process of implementation

The starting point was the training and education programme which the outside

consultants ran off site for the General Manager and his executive management team

Subsequently, a group of internally selected facilitators were taken off site and

undertook a facilitation workshop in addition to attending the training and education

programme. This group of facilitators with the assistance of the outside consultants

were given the task and considerable autonomy in selecting the process by which this

training and education was cascaded down the organization. Toward the end of 1992,

the General Manager, the executive team and the facilitators had completed the

programme. A steering committee was set up consisting of an equal number of

facilitators and members from the executive team whose task was to launch and oversee

the 'Working Together to Win' intervention.

The intervention itself was launched in a blaze of publicity in January 1993. To

stimulate interest, a variety of poster and publicity campaigns were launched. The

objectives of the intervention were communicated to the union representatives and the

entire work force. The group of facilitators ran the training and education programme

throughout the managerial/supervisory hierarchy with one day follow up work shops.

These managers and supervisors were then responsible for training their subordinates.

This cascading of the training process is similar to that conducted at Xerox whereby

training began at the top of the organization and a manger, once trained, was

responsible for training his/her immediate subordinates (Ross, 1994).

By the end of March 1993, all managers and supervisors had completed the

programme. It was now up to the individual managers to train their direct subordinates

and involve them in the intervention. Overall, the feedback given to the internal

39

facilitators on the programme was generally positive. The role of the facilitators and

steering committee was to guide and oversee first the cascading of the training and

second the foiniation of teams. Regarding the latter, a manager and his/her

subordinates, in theory, would set up a WTTW team in which a facilitator would be

present to assist in the process of objective setting, developing action plans and so

forth. The subordinates, if they were supervisors, would subsequently set up their own

teams in their work area. Thus, there would be a 'linking pin' between different

hierarchical teams facilitating the implementation of improvements. For example, a

manager would participate in a team (cross functional or with his/her superiors) which

may suggest that certain improvements be made in a manager's work area. These

suggestions would be subsequently taken to the manager's own team (in his/her work

area) for discussion and subsequent implementation.

It was assumed that after completing the programme, managers and supervisors would

actively cascade the training and set up teams. In practice, some managers did while

others did not. Some speculations can be offered as to why there was 'resistance' by

some individuals in supervisory positions to training their subordinates and involving

them in the intervention. First, a number of individuals in question (prior to the

reorganization of production) were foremen. Thus, while their job title and job content

had changed, their attitude to their job as one of directing and controlling may have

remained unchanged. The tenets of the intervention with particular reference to

managing styles may have been perceived as undermining their authority and

responsibility. Second, it is possible that the perceived demands of the job constrained

the extent to which managers could cascade the intervention. The resistance by a

number of managers was underestimated and consequently there was no pre planned

strategy for dealing with it. This lack of commitment on the part of some managers

resulted in a very uneven cascading process. By the time of the second round

40

questionnaire in September/October 1993, in some areas, the training had not only

cascaded to the bottom of the organization but also teams had been set up. In contrast,

in other areas, employees were still in the dark regarding the intervention apart from the

information that had been communicated to them when the intervention was launched

nine months previously.

The steering committee decided that in dealing with the reluctant mangers, they would

launch a 'kick start' meeting. This involved each manager presenting to the steering

committee the progress they had made in cascading the training and involving

subordinates in the intervention. This was viewed as a way to induce change by

applying pressure to these managers for action. For some individual managers, this had

some effect. While it did not lead to wholehearted enthusiasm and commitment, they

did make some progress. For the core resistors, this proved ineffective in stimulating

change

In the autumn of 1993, in view of the pockets of resistance at managerial levels, it was

decided that progress in the intervention was to become an integral part of each

manager's annual performance objectives and thus part of their performance appraisal.

Despite the attempts of one union to bring the intervention into the annual pay

negotiations, in keeping with the traditional TQM philosophy, there was no financial

incentive offered to employees for their participation. After some debate, the steering

committee decided against compulsory participation at employee levels. However, all

new employees, would be required as part of their job to participate in the intervention. 3

3 During 1994, the site took over the manufacturing of a new product from a different site and consequently hired a new group of employees (approximately 70). This new group of employees were informed of the intervention and a condition of their employment at the site was their participation in the intervention.

41

This study was conducted early in the change process and consequently captured the

uneven cascading of the intervention. Around the time of the second round of data

collection for this research, there was a noticeable attempt on directly targeting

managerial behaviour as a means to ensure that the intervention was cascaded

throughout the site. Furthermore, this shift to focusing on behaviour continued with

new employees (as mentioned above) joining the site being required to participate in

the intervention.

2.6 Summary

This chapter provided a descriptive account of the organizational context in which this

study took place. This involved providing an historical portrayal of the changes that

occurred at the site prior to the implementation of the TQM intervention. As a

precursor to describing its content, the rationale given by management for

implementing the intervention was presented. The final element of this chapter focused

on the process of implementation.

Having provided a description of the context of this study, the subsequent chapter

complements this by presenting the research methodology employed. The focus of the

next chapter is on the approach taken to gathering the data and the logic underlying the

evaluation of the intervention.

42

Chapter 3: Research Methodology

3.1 Introduction

The opening chapter provided a summary and critique of the literature thereby

highlighting the theoretical and empirical gaps to date in TQM. From this, the chapter

raised a number of research questions that the thesis sets out to address. Building upon

the descriptive account of the site and the intervention in the previous chapter, this

chapter addresses how the research questions may be answered. Given the types of

research issues being examined, what is the most appropriate method of investigation?

This chapter presents a detailed overview of the research methodology employed in this

study. This involves a discussion of two related issues: the approach taken to the

gathering of the data and the logic underlying the evaluation of the intervention.

Therefore, the aim of this chapter is to integrate the research questions, research

methodology and data analysis so that the results can be presented in the subsequent

chapters.

We start by re-examining in greater detail the research questions. Subsequently, in

view of the paucity of evaluation studies in TQM, a number of studies evaluating other

organizational change interventions are discussed with reference to the types of

research methodologies adopted. Thus, given the research questions and the

methodologies employed in previous evaluation oriented research, what is the ideal

research methodology for this study? By outlining an ideal research design, the actual

research methodology employed can be assessed against a benchmark. This will

highlight the limitations of the actual research design and also assess the feasibility of

43

some aspects of the ideal design. The final issue that is addressed in this chapter is the

statistical procedures used to analyse the data.

3.2 Research questions

A key issue in the selection of a research design is its appropriateness in addressing

research questions. Therefore, the research questions are first examined in greater

detail as a basis for discussing the research methodology in the remainder of the

chapter.

The previous chapter discussed the nature of possible TQM interventions and provided

a descriptive account of the intervention which is the focus of this investigation. The

key research question is: what is the impact, if any, of this intervention on work

attitudes? As mentioned in chapter 1, this begs a further question as to how one

assesses the impact of the intervention. One of the first problems faced in following a

`scientific' approach to conducting research is the derivation of hypotheses from the

theory. This is particularly relevant in the case of TQM in view of the absence of

explicit theories or conceptual frameworks to guide an evaluation. In chapter 1, a

strong theoretical rationale was presented for examining the antecedents of teamwork

and continuous improvement. Here, the emphasis shifts to the methodological reasons

thereby highlighting the logic underlying this evaluation.

The research questions and thus the logic underlying the evaluation of the intervention

are depicted in Figure 3.1. First, stage A, addresses the initial research question on the

antecedents of the key elements of TQM in this study: teamwork and continuous

improvement. Stage B1 examines the direct effects of the intervention on teamwork

and continuous improvement, while stage B2 + A relates to the indirect impact of the

44

intervention on the two outcomes. Finally, stage C, sheds light on the third research

question regarding the antecedents of participation in and assessment of the TQM

intervention. The logic of this evaluation is now discussed in greater detail.

Figure 3.1: Logic underlying the evaluation of a TQM intervention

TQM intervention (B2)

Intervening (A)

TQM (participation & assessment) variables — — > outcomes

(antecedents of TQM outcomes)

(C) /

(B1) Predictors

As mentioned in chapter 1, we have little knowledge of how a TQM intervention would

affect the outcomes of teamwork and continuous improvement. For example, does one

assume that if a TQM intervention affects continuous improvement, it will do so

directly? For example, individuals who participate in the intervention will subsequently

become more committed to improvement. While this indeed may be the case, it is also

possible that the intervention may have an indirect effect on teamwork and continuous

improvement. The intervention may affect, for example, some of the antecedents of

teamwork which in turn affect teamwork. Hence, the intervention may have an impact

in several ways: a direct effect, an indirect effect and a combination of both.

Therefore, this evaluation examines the total (direct and indirect) impact of the

intervention on teamwork and continuous improvement. The direct effects of the

intervention are examined in Stage B1 depicted in Figure 3.1. In assessing the direct

effects of the intervention on the outcomes, it is important to control for other potential

effects. Hence, in examining the direct effects of the intervention on teamwork, for

example, the effects of the hypothesized antecedents of teamwork are controlled for.

45

Here, the hypothesized antecedents take on the role of control variables. This permits a

more rigorous determination of the direct effects of the intervention on the two

outcomes.

The indirect effects of the intervention are examined by looking at the effects of the

intervention on the antecedents (B2) which in turn affect the outcome (A). Here, the

antecedents provide a way to capture some of the indirect effects of the intervention on

teamwork and continuous improvement. An example of an indirect effect may be that

the intervention has an impact on an employee's perception of their supervisor's

commitment to improvement which subsequently has an effect on an employee's own

commitment to improvement. This does not preclude the intervention also having a

direct effect on an employee's commitment to improvement via their own participation

in the intervention.

Therefore, the logic underlying this evaluation is that the total impact of the

intervention must be assessed. The rationale for assessing both the direct and indirect

effects of the intervention on outcome(s) is twofold. First, relying on either the direct

and indirect effects alone as a basis for evaluation may lead to an incorrect conclusion.

Had Peccei and Wood (1994, and Wood and Peccei, 1995) in their study focused

exclusively on the direct effects of the TQM intervention on quality consciousness as a

basis for their evaluation, they would have concluded that the intervention did not have

a significant effect. Rather, in using both the direct and indirect effects, the authors

concluded that the TQM intervention did indeed have a significant total effect on

quality consciousness. This approach has been previously used to examine, for

example, the total impact of a range of antecedents on innovative behaviour (Scott and

Bruce, 1994), absenteeism (Deery et al., 1995), and the impact of two different

interventions on organizational commitment (Guest and Peccei, 1994).

46

The second rationale for including both direct and indirect effects as a basis for

evaluation is to capture the unintended consequences of the intervention. This is

particularly relevant to evaluations of TQM interventions given that few have been

conducted and consequently, there is a lack of knowledge as to extent of its impact as

well as the mechanisms by which it has an impact.

To summarise, in order to evaluate the impact of the TQM intervention on teamwork

and continuous improvement, the first research question asks what are the antecedents

of teamwork and continuous improvement? This provides a basis for addressing the

second research question: what is the total impact of the intervention on teamwork and

continuous improvement?

In examining the effects of the intervention, we are interested in the effects of

individuals' participation in and assessment of the intervention. Therefore, a more

accurate reflection of the research question is: what is the total impact of individuals'

participation in and assessment of the intervention on teamwork and continuous

improvement? Underlying this are two further questions. First, is participation in the

intervention a sufficient condition for affecting change in teamwork and continuous

improvement? And comparatively, how important is an individual's assessment or

judgement of the intervention in affecting change in teamwork and continuous

improvement?

To the extent that participation in and assessment of the intervention has an impact on

teamwork and continuous improvement, what factors may affect an individual's

participation in and assessment of the intervention? While this question is not

evaluation driven per se, it needs to be addressed in order to provide a more complete

47

picture of the potential impact of the TQM intervention. As a change intervention,

TQM is organizationwide and in theory, relies on voluntary participation from

employees. Given that some employees may choose not to participate and that

employees may differ in how they assess the intervention, this has a potentially strong

effect on the outcomes of the intervention. In addition, this question has policy

implications for organizations embarking on TQM.

In summary, the thesis addresses three questions. First, what are the antecedents of

teamwork and continuous improvement? What is the total impact of individuals'

participation in and assessment of the intervention on teamwork and continuous

improvement? Finally, what factors affect an individual's participation in and

assessment of the intervention?

Having outlined the three research questions, the logical next step that warrants

attention is the research method by which these questions can be appropriately

answered. Does the existing research on TQM offer any insights as to an appropriate

methodology? What types of research designs have been employed to evaluate other

organizational change interventions? How appropriate and feasible are they for

addressing the present research questions?

3.3 Research approaches: choice?

What is the most appropriate research design to adopt given the nature of the research

questions posed? The majority of the research investigating TQM in organizations has

adopted a case study approach which has tended toward ex post facto analysis. As

mentioned in the opening chapter, there is a scarcity of evaluation driven research in

TQM (exceptions would include Guest and Peccei, 1994; Peccei and Wood, 1994;

48

Wood and Peccei, 1995) and the research questions of interest here are novel in the

sense that they have not received systematic attention to date.

Before reviewing the more common research designs employed in evaluation studies, it

is important to present a rationale for rejecting other research designs. The use of a

qualitative research approach does not facilitate the empirical testing of theoretically

constructed models as it emphasizes the subjects' interpretations or understandings of

social reality. This does not preclude the use of qualitative methods as a precursor to,

for example, the design of a questionnaire. Overall, a primarily qualitatively driven

research design may be more appropriate to investigating individuals' interpretation of

what TQM is, the reasoning behind their thinking on TQM and the importance they

attach to TQM within the context of their role or job in an organization. This

qualitative approach is quite prominent in examining a variety of issues relating to

TQM (see for example, McArdle et al., 1995; Rees, 1995; Dawson, 1995; Wilkinson et

al., 1991).

A common approach to empirically testing theoretical models is the use of a

quantitative approach consisting of a questionnaire. Of central importance is how the

questionnaire is utilized. The causality problem of cross sectional surveys has been

well documented (Bryman, 1989). In this study, the ability to attribute cause and effect

goes to the core of the research questions. A questionnaire at a single juncture would

not provide a strong basis, for example, for ascertaining the antecedents (causes) of

employee assessment of a TQM intervention (effect). Pursuing this, if organizational

commitment was found to be related to the assessment of the intervention, what is the

direction of influence? Organizational commitment may have an effect on assessment

or assessment may subsequently affect organizational commitment. Both

interpretations are equally plausible. Cross sectional data does not provide an adequate

49

basis for inferring causal connections unless the logic of causal order can be

reconstructed. While a cross sectional questionnaire may include retrospective data,

attempting to evaluate the impact of a change program using this method may be

susceptible to a host of contamination effects.

Therefore, these two research approaches; a qualitative research design and a cross

sectional quantitative approach, have been deemed inappropriate to address the research

questions in this study. From here, the next step is to review the methodologies

employed in other evaluation studies of change interventions.

3.3.1 Evaluation: organizational change interventions

Probably the most frequent research design employed to study the implications of

interventions in organizations is a quasi-experimental design or a variation of this. For

example, in assessing the impact of Quality Circles (QCs), a number of researchers

have adopted a pretest-posttest non equivalent control group design (Griffin, 1988;

Marks et al., 1986). Similarly, this type of design has been used to investigate the

effects of work redesign or changes to the content of jobs (Hackman et al., 1978; Wall

et al., 1986). In addition, studies examining the effects of participation in decision

making have adopted a variation of the quasi-experimental research design (Coch and

French, 1948; Lawler and Hackman, 1969). Guest and Peccei (1994) adopt a quasi-

experimental design with multiple equivalent groups (using three treatment groups) to

investigate the impact of two different interventions on organizational commitment.

Therefore, it would seem that a quasi-experimental research design is appropriate for

investigating organizational change. This type of design facilitates the establishment of

causality; does a particular intervention (cause) lead to certain outcomes (effect)? Two

50

key elements of this approach are central to the attribution of cause and effect. First,

there is a longitudinal dimension; that is, measurement before and after the

intervention. This permits an examination of the change that has occurred in the period

between the measurements. If change has occurred, is this a result of the intervention?

The second element of a quasi-experimental design; the establishment of a control

group helps determine the explanation for the change. If the change is only visible in

the experimental group which had the intervention and no apparent change has occurred

in the control group, one is in a better position to conclude that the intervention was the

cause of the change in the experimental group (assuming that the two groups were

comparable on all the relevant dimensions to begin with).

3.3.2 Ideal research design

At face value, a quasi-experimental design would seem to satisfy the requirements of:

(a) assessing change over time and (b) ascertaining the effect of the intervention on that

change. Using this as a basis, the following are the important characteristics of an ideal

research design. First, it would be necessary to examine the short and longer term

effects of the intervention. This would involve several waves of post intervention

measurement. The rationale for this is the view that TQM requires a longer time to

show the full benefits. In addition, short term observed changes may not be mirrored in

the longer term. For example, Griffin (1988) found that the positive effects of QCs

were apparent in the initial period but subsequently declined in the longer term. On the

contrary, Hand et al. (1973) found no significant change (in the short term) in

managerial attitudes and behaviour as a result of human relations training but

significant change was found in the longer term. The notion of short term and long

ten I change is wide open to interpretation and one researcher's idea of short term is

51

another researcher's long term. Clearly, these time orientations must be viewed in the

context of the intervention and when changes could realistically be expected to occur.

In the particular case of TQM due to its organizationwide emphasis, it would be

necessary to have a control or comparison group outside the unit implementing the

TQM intervention. This is to avoid contamination of the control group by what is

happening elsewhere in the organization. Therefore, in ideal terms, what is necessary is

a comparison group (this could be another site within the same organization) that is

comparable in all important respects. In other words, the only differentiating

characteristic between the two groups is that one group is implementing a TQM

intervention and the other is not.

Bryman (1989) argues that a number of experimentation principles are transgressed in

field experimentation due to a lack of control of the researcher over events and

experimental arrangements. For example, Jenkins and Lawler (1981) in their study of a

single organization (with no control group) examined the effects of a new payment

scheme. Two changes were introduced: a participatively designed pay plan and a pay

increase between the pre and post measurements. In this study, the introduction of

more than one change rendered it difficult to disentangle the effects on the outcome.

Thus, to a greater degree than the change interventions outlined earlier, the introduction

of TQM may potentially involve changes to the organizational structure, and processes

as well as individual attitudes and behaviour. Steel and Jennings (1992) argue that

TQM interventions do not lend themselves to conventional research strategies. They

highlight that one of the reasons is the multifaceted nature which makes it difficult to

isolate the effects of individual changes.

52

Therefore, an ideal research design would have the following elements. First, a

longitudinal element that captures both the short and longer term effects. Second, a

control group that is unaffected by the intervention. Third, the absence of other

changes that may be introduced with a TQM intervention or introduced at the same

time but unconnected to the intervention. If other changes do occur, they need to be

controlled for so that their effect on the outcome can be assessed. What may be

perceived as minor changes such as individuals changing jobs or their supervisors

changing during the implementation of the intervention may have a significant effect on

the outcomes of the intervention. For example, in this study, one of the objectives of

the intervention was to increase co-operative interaction between individuals in a work

group (i.e. team orientation). During the course of the intervention, if an individual

changed jobs and moved into a different work group, this may have an effect on their

team orientation. Thus, this type of effect needs to be controlled for in examining the

effect of the intervention on team orientation.

Are the characteristics of the ideal research design feasible? Pettigrew (1990) argues

that in- the social sciences, longitudinal research has always been in the minority. In

practice, longitudinal research faces the potential unique obstacles of continued access

for the required time period and also a lack of control over the events that occur in the

intervening period. However, this serves to make longitudinal research more difficult

rather than unfeasible. The establishment of a control group may pose a more difficult

problem in the context of TQM due to its organizationwide emphasis.

Regardless of the type of intervention being examined, the establishment of a control or

comparison group seems to be generally difficult to accomplish. Consequently, the

absence of control groups in studies of organizational change is not unusual.

Cummings, Molloy and Glen (1977) report that 23 out of the 58 work experiments

53

analysed did not have a control or comparison group. Obtaining an uncontaminated

control group within the same organization may be difficult due to the

organizationwide (or unit wide) focus of TQM. What may be even more difficult is

finding a comparable organization that is similar on all the relevant variables with the

exception of the TQM intervention.

Rossi and Freeman (1993) argue that the most severe restriction on the choice of

research design employed to evaluate interventions is whether or not the intervention in

question is being delivered to all members of the target population. The examples

previously presented of quasi-experimental research designs consisted of partial

coverage programs whereas in theory, TQM is a full coverage intervention. To

overcome the absence of control groups in full coverage programs, two strategies may

be adopted contingent upon the number of post intervention measurements. First, if the

intervention progresses slowly to full coverage (thus, particularly at the early stages

individuals will be differentially exposed to the intervention), repeated post intervention

measurements will allow the identification of processes by which the intervention

affects the individuals. If a post intervention measurement is taken only once,

depending on the speed of diffusion of coverage, a post hoc control group may be a

possibility. Lawler (1977) argues that in the absence of a control group, data should be

gathered everywhere the change is expected to impact and if the change does not affect

one area/group or if it has been less affected, a post hoc control group may be

established.

3.4 Research design adopted

In many situations, it is difficult or unfeasible to conduct research using in ideal terms

the best possible research design. Rossi and Freeman (1993) advocate the 'good

54

enough' rule in selecting research designs for evaluating change programs.

Consequently, trade-offs are made in formulating the best possible design taking into

account methodological issues, practicalities and overall feasibility. We now turn to

the research design employed in this study.

The initial focus is on the case study approach. In particular, two elements of the

research design are discussed: the longitudinal dimension and comparison group.

Following from this, the research method; that is, the questionnaires are reviewed.

Finally, the process of data collection from the point of entry into the organization is

described.

3.4.1 Case study approach

As mentioned in the previous section, the nature of TQM does not lend itself to a quasi-

experimental research design. Therefore, the research framework adopted here does not

meet the criteria of a quasi-experimental research design and may be better described as

a longitudinal case study.

A case study approach as defined by Yin (1984) allows a detailed investigation of a

particular phenomenon within the context in which it occurs. Therefore, in this study, a

case study approach permits a more in-depth examination of a TQM intervention and

places it in the context in which it occurs. This approach has been put forward as one of

the primary research designs in which quantitative and qualitative methods may be

combined. It has been argued that it is unusual for quantitative and qualitative research

methods to be allocated an equal role within the overall research design. This study is

no exception in that greater emphasis was placed on quantitative methods. To a greater

extent, the use of quantitative versus qualitative methods was contingent upon the stage

55

of the research process. For example, at the commencement stage of this study,

unstructured interviews and documentation provided contextual information with

regard to previous organizational changes and future expectations regarding the planned

TQM intervention. Subsequently, quantitative methods came to the fore in terms of

providing a baseline measure upon which to assess the extent of change. Between the

collection of the baseline questionnaire and the second questionnaire (some 15 months

later), interviews which were highly unstructured and quite ad hoc (particularly with

employees) fulfilled two roles. First, as a way of monitoring the progress of the

intervention and second, as a source of additional questions to be included in the second

questionnaire.

3.4.2 Longitudinal research

From a methodological viewpoint, a longitudinal dimension is a requirement for

evaluating organizational change interventions. The very nature and aim of evaluation

studies dictate a measurement before and after the intervention in question. While the

advantages of longitudinal research are well known and documented (Kimberly, 1979;

Pettigrew, 1990), there are some unresolved issues.

How many data collection points are necessary for a study to be longitudinal? How

long between collection points is longitudinal? This study has two collection points

reflecting pre and post intervention measurement. Pre intervention measurement

occurred six months prior to the intervention while post measurement took place nine

months after the intervention. Given this time span, another data collection point would

not have been practical and may have reduced the quality of the data obtained.' There

Campbell and Stanley (1966) argue that in studies dealing with human subjects, too frequent data collections is likely to have negative implications for the quality of data. In addition, it may not have been practical in terms of working time lost.

56

is an absence of guidelines in formulating an appropriate span of time between

measurements. In addition, of particular importance in the evaluation of organizational

change interventions, the researcher has little or no control over the change process

which may affect the timing of the post measurement. Consequently, several issues

may assist in the timing of a post intervention questionnaire. First, when could one

reasonably expect the effects to occur? This clearly depends on the phenomenon under

investigation. For example, a change in a group payment scheme may exhibit effects

much quicker than an organizationwide intervention to change culture. Second, what

guidelines do other organizational change interventions provide? Third, is the

establishment of a post hoc control group necessary? If so, this may require the

researcher to intervene earlier in the change process than otherwise necessary. Finally,

the time lag between measurements may have a substantial effect on the results. For

example, if Griffin (1988) had limited the time lag to 18 months in investigating the

effects of QCs, a different conclusion would be reached compared to a time lag of 36

months.

The guideline from other evaluation of change studies indicate an initial post change

intervention measurement between 6 months and one year. In other longitudinal

research in TQM (Peccei and Wood, 1994; Wood and Peccei, 1995), a time span of one

year was adopted. Consequently, the time lapse in this study is more common than not.

It was important in this study to intervene early in the change process so that certain

areas were largely unaffected by the change process and thus could act as a post hoc

control group. However, one could plausibly argue that any observed changes or the

assessment of the impact of the intervention is purely short term. This, is a limitation

of this study.

57

To summarise, the longitudinal dimensions consisted of a two wave before and after

study of the TQM intervention. The pre measurement occurred approximately six

months prior to the intervention at which stage none of the individuals completing the

questionnaire knew of the pending TQM intervention. Consequently, the baseline

questionnaire was not contaminated or influenced by individuals' knowledge of the

forthcoming intervention. The post measurement took place nine months after the

commencement of the intervention. This short time span is a limitation of this study

and the results may not be consistent with a longer term evaluation. Some other

evaluation studies overcome this by assessing both the short and long term effects.

However, in this study, the results need to be interpreted within the context of the time

span allowed.

3.4.3 A comparison site

While it would have been ideal to include a comparison site in this study, it was not

possible. As indicated in the previous chapter, a second site was selected based on its

progress in and different emphasis on TQM. Having gained access, the procedures

followed were similar to that of site 1 (this is discussed later). However, during the

course of the case study, there were unforeseen circumstances by way of the sale of the

site and redundancies that subsequently rendered the initial objective in using this site

unfeasible. This highlights one of the inherent risks in conducting longitudinal

research. For the purpose of this research, the role and emphasis assigned to the second

site was modified.

The primary contribution of including the data from the second site (hereafter site 2) is

in the testing of the models prior to evaluating the impact of the TQM intervention at

58

what is, hereafter referred to as site 1. Thus, the models are more rigorously tested

using two independent samples. Given, the different changes that occurred at the two

sites, this allows a test of how robust the models are in predicting the dependent

variables under different organizational conditions.

3.4.4 The questionnaires

The questionnaires used in this study are presented in Appendix 1. This includes the

time 1 and time 2 questionnaires for supervisors and employees. The aim of this

section is to discuss the design of the questionnaire in general terms due to the large

number of measures it contains. For ease of clarity, the measures are discussed in detail

in the relevant chapters.

In devising the questionnaires, three issues were given important consideration. These

are dealt with in turn. First, to facilitate comparisons between groups, it was necessary

for the questionnaire to be appropriate to a diverse range of respondents. Therefore, the

questionnaire needed to be easily completed by unsophisticated respondents.

Second, where possible, previously validated scales were used. The issue of scale or

measurement reliability is of particular importance in longitudinal studies. The relevant

issue is the distinction between unreliability in measurement (error) and legitimate

change in longitudinal studies. Therefore, scales need to exhibit high stability and yield

reproducible results. The majority of the validated scales were taken from Wan, Cook

and Wall (1979), Cook and Wall (1980) and Cook, Hepworth, Wall and Wan (1981).

These researchers specifically set out to develop and validate a variety of quality of

work life measures. Overall, evidence is provided to show that the developed scales are

psychometrically adequate and stable. However, it was not possible to rely exclusively

59

on previously validated measures. Some of the concepts used in TQM are new and had

not been operationalized when this study commenced. Thus, concepts such as quality

awareness, perceived management and supervisory commitment to quality and

commitment to improvement were created for this study. In the relevant chapters, these

measures are discussed and reliability and factor analytic data are presented.

Finally, the questionnaire needed to include a comprehensive range of measures for two

reasons. First, it was necessary to measure a relatively large number of antecedents of

the dependent variables. Second, given the few evaluation studies of TQM and

consequently, the lack of knowledge as to the impact of TQM interventions, a wide

measurement net is in a better position to capture the intended and unintended effects of

the intervention.

A pilot study was conducted on the full questionnaire at time 1 and the new questions

added to the questionnaire at time 2 on a small sample of individuals. The priority in

selecting participants for the pilot study was to include a diverse range of respondents

from machine operators to senior managers. From the results of the pilot study, the

major modification needed was the omittence from the employee questionnaire of their

perception of their immediate supervisor's subscription to Theory X and Y. This has

been successfully measured in another study (Fiman, 1973) which consisted of white

collar employees. The employees in the pilot study who worked on the shop floor

reported great difficulty in giving their perception of their supervisor's view of Theory

X and Y. Consequently, these items were dropped from the employee questionnaire.

The same items were retained in the supervisory questionnaire which asked supervisors

for their own view. In addition, an open ended question asking respondents'

interpretation of TQM was omitted due to its non response in the pilot.

60

The previously validated scales were replicated insofar as possible in this questionnaire.

This included the presentation of the items in a scale in a block prior to moving on to

the next scale. This strategy was also used with the new measures created for this

study. In terms of the ordering of the measures within the questionnaire, the more

sensitive questions relating to an individual's perception of their immediate boss or

questions relating to management were placed in later sections of the questionnaire.

3.4.5 Data collection

Prior to outlining the actual data collection method, the introduction process at the

beginning of this study is briefly described. Subsequent to access being agreed upon

with the general manager, the next step involved meeting the union representatives.

This allowed them the opportunity to raise any questions or concerns they may have

regarding the research. At the site four times a year is a communications day whereby

all employees in groups are given a forty minute presentation on relevant issues to the

site. This provided the opportunity for me to introduce myself to the entire work force,

describe in general terms the research and state my independence from management at

the site and from the overall organization. All these steps were taken to facilitate the

continued co-operation needed.

On the basis of a list of all employees by work area or functional area, a stratified

random sample of 40% was taken. This procedure was only applied to employees as it

was necessary to include all managers and supervisors (except the executive team) in

the sample. Having identified the individuals selected to participate, they were asked if

they would be willing to complete a questionnaire and informed that it was voluntary.

Supervisors, managers and engineers completed the questionnaire in a group situation

under my supervision. This provided a further opportunity for individuals to raise

61

questions and allowed me to alleviate any concerns over the use of identification

numbers for the purpose of tracking individuals over time. Most of the employees

completed the questionnaire on a one to one basis during work time. This method

although time consuming was deemed to be superior to self administered questionnaires

and also provided an opportunity to gain further information regarding the organization

and its climate. By completing the questionnaires using this method, it was felt that the

length of the questionnaires could exceed that of a self administered questionnaire.

This data collection method was replicated for the time 2 questionnaire.

3.4.6 Sample characteristics

At time 1, of the 255 individuals (40% of employees and all supervisory personnel

excluding the General Manager and the executive management team) asked to complete

the questionnaire, 241 did so yielding a response rate of 94.5%. 2 Of the 241

participants, 186 were employees and the remaining 55 were in a supervisory position.

At time 2, the participant sample was reduced to 216 primarily as a result of individuals

leaving the organization in the intervening period. Table 3.1 presents a summary of the

sample characteristics of site 1.

The participant group was 94% male, with a mean age of 47.6 years, a mean

organizational tenure of 18.2 years and a mean job tenure of 8.3 years. The majority of

the sample (35.6%) were machine operators, 22.2% craftsmen and 23.2% in a

supervisory position. The sample characteristics of site 2 are presented in Appendix 2

and differ from site 1 except for work force composition and job categories.

2 At site 2, 316 completed the questionnaire at time 1 and this was reduced to 228 at time 2 primarily as the result of redundancies occurring in the intervening period. Of the 228 participants at time 2, 183 were employees and 45 managers / managers.

62

Table 3.1: Sample characteristics of site 1

Site 1

1. Mean Age (S.D) 47.6 years (9.8) 2. Length of time in present job (S.D) 8.3 years (7.6) 3. Length of service with organization (S.D) 18.2 years (9.5) 4. Length of service at site (S.D) 16.8 years (10.4) 5. Gender

Male (% of sample) 203 (94%) Female (% of sample) 13 (6%)

6. Job Categories Operators (% of sample) 77 (35.6%) Craftsmen (% of sample) 48 (22.2%) Engineers & Material controllers (% of sample)

43 (19.0%)

Clerical/Administrative (% of sample) 17 (8.0%) Supervisors/Managers (% of sample) 51 (23.2%)

3.5 Ideal vs. actual research design: limitations

Three particular features of an ideal research design were discussed; a longitudinal

dimension, a control group and other changes that may occur concurrent with the

intervention. In an ideal research design, a number of post intervention measurements

would be taken in order to capture the process of change and also provide a short and

long term evaluation. The latter element would help assess the temporal validity of the

short term findings. Regarding the control group, the ideal situation would be a

comparable group removed from the intervention group so as to avoid the potential

effects of contamination. Finally, it would be ideal not to have other changes occurring

that may potentially have an effect on an evaluation of the outcomes of the intervention.

Briefly, the actual research design employed involved one post intervention

measurement at nine months and did not contain a 'proper' control group. These two

63

issues may be considered limitations of this study. Consequently, this study can only

evaluate the TQM intervention in the short term and the results must be interpreted with

this in mind. However, this time span did permit the establishment of a post hoc

control group; at the time of the second round questionnaire, approximately 50% of

employees were participating in the intervention thus providing a balanced variation

between participants and non participants.

Regarding other changes that may complicate the evaluation of the intervention, insofar

as it was possible, these changes were anticipated and controlled for. These changes

included a change in supervisor, a substantial change in job content and a change of

jobs which could potentially have a significant effect on an individual's team

orientation and commitment to improvement. While these type of changes were

controlled for, the extent to which they occurred were minimal For example, between

time 1 and time 2, only 8 out of the 165 employees had changed jobs, 34 employees

had a different supervisor at time 2 and 20 employees reported a substantial change in

the content of their jobs. 3 Therefore, the occurrence of other changes did not pose a

problem in this study given that the major change at the site was the intervention and

the other changes were minimal.

Perhaps, the key question is whether the research design adopted is 'good enough' to

address the research questions. Clearly, more studies than not fail to achieve in practice

the ideal research design. In reality, very few studies adopt a long term systematic

approach to the evaluation of organizational change interventions dealing with

individuals. Thus, the long term perspective adopted by Wall et al. (1986) and Griffin

(1988) of 30 and 36 months respectively is more unusual than not. Given the threats of,

for example, mortality, a lack of control over organizational events and perhaps denied

3 For the overall site sample (including supervisors), 15 out of 216 had changed jobs, 34 reported a substantial change in their jobs and 50 had a change of supervisor between time 1 and time 2.

64

access, these make long term evaluations more difficult. Therefore, while the time span

is an obvious limitation, it is not unusual.

The more serious limitation may be the absence of an a priori control group. However,

the only viable method to overcome this was to establish a post hoc control group in

terms of obtaining variation between employee participation in the intervention.

Overall, these limitations must be weighed against the strong points of the research

design and also the overall contribution of the research.

3.6 Data analysis

The aim here is to outline the statistical procedures used to examine the impact of the

intervention on the outcomes of teamwork and continuous improvement.

A two staged approach was adopted to the assessment of the impact of the intervention.

The first stage involved the construction of theoretical models containing hypothesized

antecedents of the specified outcomes of TQM. For example, in the model of the

antecedents of team orientation, theoretically justified antecedents such as trust in

colleagues and supervisor participative style were included. Consequently, the first

empirical stage required the testing of the antecedent models to assess how reasonable

they were in predicting the outcomes. As mentioned earlier, this model was tested

using two independent samples to assess the robustness of the model. Once this stage

was completed, the models were used as the basis for evaluating the impact of the

intervention.

The statistical techniques used were required to match the logic underlying the

evaluation approach and also simultaneously cope with a large number of independent

65

variables. Consequently, path analysis using Ordinary Least Square (OLS) multiple

regression was employed. Path analysis permits the estimation of the relative total

impact of a number of variables within a model. While it allows the relative impact of

variables to be estimated, path analysis cannot establish causality or confirm the

hypothesized causal model. Thus, if a theoretical causal model is incorrect regarding

the ordering of the variables, then the path analysis and the relative impact of the

variables may be misleading.

Path analysis using OLS multiple regression requires two stages. First, the intervening

variables (antecedents) are regressed separately on the intervention and control

variables. Subsequently, the dependent variable (i.e. the outcomes of TQM) is

regressed on all the variables in the model. The resulting path coefficients

(standardized beta coefficients) were used to assess the direct, indirect and total effects

of the TQM intervention on the dependent variables. For ease of clarity, the relevant

intercorrelation tables for each of the results chapter are presented in Appendix 3.

3.7 Conclusions

This chapter presented an overview of the research methodology employed in this

study. This involved outlining an ideal research design based on the research questions

and subsequently using this to assess the limitations of the actual research design

employed. The most serious limitation of the research design employed is the absence

of an a priori control group. However, the naturally occurring variation in terms of

employee participation in the intervention allowed the establishment of a post hoc

control group.

66

The underlying logic of the approach adopted to evaluate the intervention was outlined.

The initial stage involved examining the antecedents of the TQM outcomes. As the

subsequent chapter will show, theoretical models containing hypothesized antecedents

of the dependent variables are constructed and empirically tested using two independent

samples. Subsequently, these models are used as a basis to evaluate the total impact of

the intervention. The statistical procedure employed was path analysis using OLS

multiple regression. This permitted the relative impact of each of the TQM

intervention variables to be estimated.

This chapter and the previous two chapters have raised the research questions, provided

a descriptive account of the organization and the TQM intervention and discussed the

research methodology employed in this study. The subsequent chapters present the

findings. This commences with the antecedents of team orientation in the next chapter

as a precursor to evaluating the impact of the intervention on team orientation in

chapter 5.

67

Chapter 4: The Antecedents of Team Orientation

4.1 Introduction

The notion of teamwork is ubiquitous in the literature on TQM. For many contributors,

teamwork, in addition to continuous improvement and customer satisfaction

characterize TQM as a philosophy (Dean and Bowen, 1994; Sitkin, Sutcliffe and

Schroeder, 1994). Although it is used pervasively, there is some ambiguity as to what

teamwork involves and who it is directed to. Adopting a broad criterion of

applicability, the notion of teamwork applies to all organizational members. It may be

viewed as having four facets: within a natural work group, between different functional

departments, between managers and employees and finally, between organizations;

customers and suppliers. While teamwork may apply to these different groups, there is

a common underlying theme; a willingness to co-operate.

Between customers and suppliers, teamwork manifests itself in co-operative efforts to

achieve mutual benefits in terms of synergy and loyalty (Dean and Bowen, 1994).

Cross functional teamwork is based on the optimization of the whole system rather than

the attainment of functional outcomes. Functional barriers are dismantled through the

creation of a cross functional teamwork culture via the internal customer principle.

Teamwork between managers and employees is based on co-operation: employees co-

operate by contributing to continuous improvement; highlighting problems that prevent

them from doing quality work and suggesting ways that their own work, the work of

their group or the activities of the organization may be improved. Managers reciprocate

by facilitating employee contributions to improvement in terms of providing training,

setting up mechanisms by which suggestions are reviewed and action taken. The focus

of this chapter is on teamwork within a natural work group and it is to this we now turn.

68

What does teamwork mean when it is applied to employees? There seem to be two

interpretations that are evidenced in the literature although a distinction is not always

made explicit. The first interpretation focuses on work restructuring and the second

emphasises an individual's orientation toward the work group. The first perspective is

probably better reflected under the title of team based working. This involves a

restructuring of work away from individually designed tasks to group based tasks. In

manufacturing organization, this may be known as a "cell" based structure where

employees work in teams which are themselves mini organizations grouped around a

product or component. The team takes on the responsibility for the product from its

manufacture to its delivery to the internal or external customer. Within this structure,

there are great variations as to the extent of responsibilities given to the team with self

managing teams as the ideal structure (Juran, 1989). This is analogous to the

autonomous work groups (AWGs) advocated by and stemming from the work of the

socio-technical systems theorists (Emery and Trist, 1969; Rice, 1958; Trist and

B amforth, 1951).

However, it is debatable whether team based working in the context of TQM parallels

the self determination of AWGs in terms of the range of responsibilities given to the

group (e.g collective participation in recruitment and an absence of direct supervision;

Wall et al., 1986). What may be more common is that the team is given responsibility

for issues directly related to quality and improvement issues. Thus, the team is

allocated responsibility for quality (through the self inspection principle) and for

improving the effectiveness and productivity of the work group (Sewell and Wilkinson,

1992). As Wilkinson (1992) states, "...responsibilities should be team rather than

individual ones, since this makes mistakes and risk taking easier for individuals to

bear" (p 327).

69

This perspective on teamwork, in essence, views it as a structure; a method of

structuring work around a group of individuals. As such, this was implemented at the

site prior to the TQM intervention in the form of a cell based structure. If work is

restructured around a team creating interdependence of individual tasks and

consequently, highlighting the importance of co-operative activity, does this mean that

individuals will necessarily adopt a team orientation? This leads to the second

perspective on teamwork which emphasizes teamwork as an attitude or value and may

be more appropriately titled team orientation. The latter perspective was one of the

objectives of the intervention and this was consistent with the "cultural" emphasis of

implementing a 'soft' TQM intervention.

Bushe (1988) describes a total quality culture in terms of norms, values and reward

procedures that emphasize holistic behaviour oriented toward co-operation with fellow

organizational members. Waldman (1994) argues that it may be difficult to separate

work performance in terms of formal role requirements from organizational citizenship

behaviour in the context of TQM. He asserts that in a total quality culture, work

performance would include taking initiatives beyond the call of duty, sharing

information with and helping co-workers. Blackburn and Rosen (1993) in reviewing

TQM and HRM of the Baldrige Award winners contrast traditional versus TQM

paradigms toward organizational culture and HRM practices. They highlight a shift in

cultural emphasis from individualism to collective efforts. Anecdotal evidence is

provided that successful TQM companies rely on co-worker assessment of an

individual's team orientation as one criteria in measuring an individual's contribution to

teamwork. Along a similar vein, Drummond and Chell (1992) note that all employees

"will need to adopt holistic approaches to work based on co-operation" (p5).

70

In other words, this second perspective on teamwork provides insight into the social

system; it relates to the interaction of team members and an individual's orientation

toward the team. Here, it is the second perspective; team orientation, that is the focus

of investigation of this chapter.

Implicit in the mainstream literature is a positive view of the collaborative efforts and

co-operative interaction within the team. In other words, the notion of team orientation

is viewed as beneficial to the achievement of the team's objectives as defined by

management. Boje and Winsor (1993) argue that peer pressure is a more compelling

method of worker control than hierarchical control. Workers are made to feel a sense

of obligation to other members. This assumes that the norms and values of the team are

congruent with the organization. This may not necessarily be true. It has previously

been shown that workers can negatively deviate from organization-defined norms. In

other words, individuals may co-operate with and exert effort for other team members

for reasons other than the effective functioning of the group as defined by management.

Instances such as restriction of output could be interpreted as a willingness by an

individual to exert effort on behalf of other members (to help maximise their earnings).

Roy (1952), Burawoy (1979) and Roy (1969) present evidence of "goldbricking" and

"making out" where workers co-operated with each other to improve their own

conditions.

Following the thinking of TQM, this chapter assumes that team orientation assists the

functioning of the team and is congruent with the objectives of management. However,

it is acknowledged that team orientation may exist for other reasons but this is not

explored here.

71

As mentioned in chapter 3, it is important in an evaluation of the impact of a TQM

intervention to examine the direct and indirect effects of the intervention on, in this

case, team orientation. Consequently, this chapter empirically tests a model containing

hypothesized antecedents of team orientation which may be subsequently used to

assess the full effects of the intervention on team orientation. As a starting point, the

chapter begins by examining the definition of team orientation employed here.

Subsequently, the hypothesized model containing the antecedents of team orientation is

outlined. The measures used and the analysis procedures adopted are described prior to

empirically testing the hypothesized model of team orientation on two independent

samples (site 1 and site 2). Overall, the objective of this chapter is to assess how

reasonable the hypothesized model is as a basis for evaluating the impact of the

intervention on team orientation at site 1 in the subsequent chapter.

4.2 Team orientation

This section presents a definition of team orientation, and outlines the model containing

the hypothesized antecedents of team orientation to be tested using the data from the

two sites.

4.2.1 Definition of team orientation

The concept of team orientation can be viewed and interpreted as having two distinct

dimensions. The first dimension taps an individual's orientation toward the team /

work group while the second dimension taps the degree to which a team orientation is

reinforced by other members of the team or work group. Neither dimension on its own

fully captures team orientation. If one were to rely exclusively on measuring team

orientation from the first perspective, one may get a distorted view of team orientation.

72

For example, an individual may feel part of a work group and may be willing to exert

effort for the work group but within the work group, this is neither valued nor

encouraged. Consequently, while an individual may be oriented toward the team, other

members may not encourage team orientation. In order to obtain a more complete view

of team orientation, a measure would have to include an individual's orientation toward

the team as well as his or her perception of whether this team orientation is encouraged

by other team members. The definition used here includes both components captured

by the following questions in the questionnaire:

- I feel I am really part of my work group

- I am willing to put myself out to help my work group

- The people in my work group encourage each other to work as a team

The first question taps an affective dimension of an individual's orientation toward the

work group. This type of identification has been previously used in a study of

intergroup relations (Brown and Williams, 1984). However, strong identification with

the work group is arguably not sufficient in itself to tap the importance an individual

attaches to his/her team orientation. The rationale is simply that an individual may feel

part of his/her work group but may not be willing to put himself/herself out to help the

team. Consequently, what is needed is a behavioural element; that is, whether

individuals are willing to put themselves out to help the work group. This is akin to

what Waldman (1994) argues is one element of an individual's work performance in

the context of TQM. Finally, the last question taps the degree to which individuals

perceive a team focus to be encouraged by other members in the group.

73

4.2.2 The antecedents of team orientation

The hypothesized model containing the antecedents of team orientation is depicted in

Figure 4.1. Team orientation is viewed as the outcome of four categories of predictors:

individual, colleagues, leader and climate. Each category is discussed in turn beginning

with the individual category.

Individual and team orientation This category contains one predictor; an

individual's quality awareness which is hypothesized to have a positive influence on an

individual's team orientation. In other words, individuals who have heightened

awareness of the importance of quality for the organization and their work group are

more likely to exert effort on behalf of their work group. The knowledge of the

consequences of an individual's actions as well as the interdependence of individual

actions in relation to quality may provide a uniting goal for members and consequently

affect their orientation toward the team.

74

FIG

URE

4.1

: Hyp

othe

size

d m

o del

of t

he a

ntec

eden

ts o

f tea

m o

rient

atio

n

Qua

lity

awar

enes

s

Satis

fact

ion

with

colle

ague

s

Tru

st in

col

leag

ues

Team

orie

ntat

ion

75

Colleagues and team orientation This category consists of two predictors that tap

an individual's attitude toward the colleagues they work with. These include an

individual's satisfaction with their colleagues and trust in their colleagues. Individuals

who are more satisfied with their colleagues, and who have greater trust in their

colleagues are hypothesized to be more likely to have a positive team orientation and

to view it as being encouraged within the work group.

"Perhaps there is no single variable which so thoroughly influences interpersonal and

group behaviour as does trust" (Golembiewski and McCollide, 1975, p131 ). In terms

of the relationship between trust and co-operation, the direction of causation is subject

to debate. Deutsch (1962) argues that trust is the central prerequisite of co-operation

while others (Tedeschi, Hiester and Gahagan, 1969; Boyle and Bovacich, 1970) provide

evidence that there is reciprocity of influence between trust and co-operation. From

this perspective, it is argued that a co-operative environment will induce trust in

addition to trust facilitating co-operation.

In this study, trust in colleagues taps two dimensions: a belief that co-workers would

assist the individual should the need arise and; an individual's confidence in the ability

of his/her co-workers. Thus, if an individual believes his/her co-workers will help them

out, they, in exchange will be more likely to put themselves out for the work group and

perceive others, through their helping behaviour or perceived intention to do so, as

encouraging teamwork. It is plausible that the relationship between trust in colleagues

and team orientation is reciprocal. It is possible to test the nature of this relationship

between trust in colleagues and team orientation (albeit a weak test) using cross lagged

regressions. Thus, if trust in colleagues at time 1 has a significant effect on team

orientation at time 2 and team orientation at time 1 does not have a significant effect on

76

trust in colleagues at time 2, this would indicate that the relationship was not reciprocal

but rather suggest that trust in colleagues is an antecedent of team orientation. The

results of the cross lagged regressions' for site 1 and 2 are presented in Appendix 4 and

5 and support treating trust in colleagues as an antecedent of team orientation. This

would be consistent with Zand's (1972) finding that trust was related to affective

attachment. Thus, individuals who trust their colleagues are more likely to identify

with their work group.

Satisfaction with colleagues is hypothesized to have a positive effect on team

orientation. The more an individual likes his/her colleagues, the greater the likelihood

that they will feel part of the group and be willing to put themselves out to help the

group. Once again, it is possible that the relationship is reciprocal; the more an

individual likes his/her colleagues, the greater the team orientation, and the more a team

orientation is encouraged by other members, the greater an individual's satisfaction

with colleagues. As in the case of trust in colleagues, cross lagged regressions were

conducted to shed light on the relationship between satisfaction with colleagues and

team orientation. If the relationship was reciprocal, one would expect to find

satisfaction with colleagues at time 1 having a significant effect on team orientation at

time 2 and team orientation at time 1 having a significant effect on satisfaction with

colleagues at time 2. The results (Appendix 4 and 5) indicate that team orientation at

time 1 did not have a significant effect on satisfaction with colleagues at time 2 thereby

providing support for treating satisfaction with colleagues as a predictor of team

orientation.

1 Each of the hypothesized predictors at time 2 was regressed separately on all other variables in the model (including the predictor) measured at time 1.

77

Leader and team orientation. The behaviour and attitude of an individual's

immediate boss and of management more generally are hypothesized to positively

affect team orientation. More specifically, team orientation is hypothesized to be

positively affected by the participative style of the immediate supervisor. Leadership

has been conceptualized as a process whereby a leader seeks to influence followers'

beliefs and values resulting in behavioural changes (Yuki, 1989). In terms of Lewin's

(1943) psychological proximity, the immediate boss, as he/she is more proximal to the

individual is in a position to shape an individual's behaviour and the interaction of

individuals within the work group. In other words, the participative style of the leader

who facilitates, values and reinforces a team emphasis is likely to shape an individual's

team orientation and the degree to which it is encouraged within the group. This would

be consistent with Human Relations theory which argues that leaders can develop their

members into a working team with high loyalty using participation (Likert, 1961).

In addition, perceived management commitment to quality is hypothesized to affect

team orientation. In the TQM literature, critical importance is placed on management

commitment to quality values (Deming, 1986 and ; Juran, 1989; Oakland, 1989; Dale

and Cooper, 1992) as it portrays to organizational members what is important and

signals the attitudes and behaviours that are valued. Thus, a committed management

team would set an example of co-operative behaviour within their own hierarchy,

communicate and reinforce team values to the lower levels and support co-operative

efforts and interaction.

Climate and team orientation. The final factor of an improvement in

commitment to quality in general at the site reflects an individual's perception of a

change in the cultural emphasis or value given to quality within the organization and

associated values and practices such as teamwork. This is hypothesized to have an

78

impact on shaping an individual's team orientation and perception of teamwork within

their work group. Previous research has highlighted the importance of climate

perceptions in affecting individual attitudes and behaviour (Scott and Bruce, 1994;

Schneider, 1983a).

Within the categories outlined, the predictors may be labelled as either directly tapping

quality related attitudes and perceptions or not. This is important in light of the lack of

theoretical grounding underlying TQM. Consequently, the predictors that explicit tap

quality attitudes and perceptions may best be treated as exploratory predictors. One

could expect a TQM intervention to affect quality awareness, perceived management

commitment to quality and quality climate. However, whether these factors in turn

affect team orientation is open to empirical testing. Unlike the other hypothesized

predictors, these "quality" predictors are not grounded with a strong theoretical

rationale and thus need to be considered more as exploratory predictors.

4.3 Measures

As discussed in chapter 3, where it was possible, measures which had been previously

validated were used. In the forthcoming discussion of the measures, the source of the

previously validated measures is highlighted. Unless otherwise indicated, responses to

the questionnaire items were scored on a seven point Likert scale measuring a

respondents' satisfaction/dissatisfaction or agreement/disagreement with an item. The

change variables were computed by subtracting the Time 1 scores from the Time 2

scores to assess the amount of change that had occurred in each of the measures over

time. For the purpose of clarity, the alpha coefficients are given for the combined site 1

and 2 sample at time 1 and 2. 2

2 The alpha coefficients for the individual site samples do not vary much from the combined sample.

79

Team orientation This is a three item scale tapping team orientation. It consists of

two dimensions: (i) the degree to which an individual is oriented toward the work group

and (ii) the degree to which a team emphasis is perceived to be reinforced by other

members in the group. The alpha coefficient for this scale was .76 at time 1 and .68 at

time 2.

Quality awareness This variable was measured using a four item scale constructed

to tap respondents' awareness of the importance of quality. The items tap two

dimensions; (i) awareness of the importance of quality work and continuous

improvement for the organization and (ii) specific ideas to improve the quality of work

in the work area. The alpha coefficients for this scale was .60 at time 1 and .54 at time

2.

Trust in colleagues This was measured using Cook and Wall's (1980) six item

interpersonal trust in peers scale. This scale taps two dimensions of interpersonal trust;

(i) faith in the trustworthy intentions of others and (ii) confidence in the ability of

others. The results of the factor analysis provide support for the factorial independence

of this measure of trust in colleagues and team orientation (see Appendix 6 and 7). The

scale exhibited high levels of reliability at time 1 and time 2, .86 and .85 respectively.

Satisfaction with colleagues This is a single item measure taken from Warr,

Cook and Wall's (1979) overall job satisfaction scale that taps respondents' satisfaction

with their colleagues. As this is a single item measure a reliability coefficient was not

calculated.

Supervisor participative style This variable refers to the extent to which

respondents' perceive their immediate boss as being participative and supportive in their

80

behaviour. The items tap two dimensions: (i) the degree to which the immediate boss

encourages participation and facilitates the group working together and (ii) the extent to

which the immediate boss supports the individual in their work and makes use of their

work knowledge and abilities. The alpha coefficient for this seven item scale was .86 at

time 1 and .88 at time 2.

Management commitment to quality This variable refers to perceived

management commitment to quality. The item taps two dimensions of the concept; (i)

the extent to which management is perceived to set an example of quality behaviour in

their own work and (ii) the extent to which management is perceived to facilitate

employees doing quality work and provide support for quality improvements. This six

item scale exhibited high levels of reliability, .84 and .89 at time 1 and time 2

respectively.

Improvement in commitment to quality This five item scale measured at time 2

taps respondents' assessment of whether people in general in the organization had

become more committed to quality during the previous year. This measure was

designed to tap greater perceived commitment to quality in the organization as a whole

and, as such, is conceptually distinct from the measure of perceived management

commitment to quality. The results of factor analysis provide support for the factorial

independence of these two measures (see Appendix 8 and 9) The alpha coefficient for

this scale at time 2 was .89.

Control variables These included respondents' age, gender, organizational tenure,

length of time in present job and job title. Three dummy variables were created for job

title (direct production employees, qualified employees connected with production and

administrative employees) with direct production employees being used as the

81

reference category. Three additional control variables were added at time 2; whether

respondents' had the same immediate boss as they did when they completed the first

questionnaire; if the content of their job had changed substantially and if they changed

their jobs between time 1 and time 2.

4.4 Analysis procedures

The model outlined in Figure 4.1 assumes that team orientation is endogenous in

relation to the other variables and consequently that it is appropriate to treat team

orientation as the dependent variable. It is possible to test the endogeneity assumption

through the use of cross lagged regressions. This has previously been discussed for the

specific variables of trust in colleagues and satisfaction with colleagues. Here, this

analysis is extended to all the other hypothesized predictors of team orientation. This

involves regressing the individual predictor variables measured at time 2 on team

orientation measured at time 1. The results are shown in Appendix 4 for site 1 and

Appendix 5 for site 2. At site 1, team orientation at time 1 did not have a significant

effect on any of the predictor variables at time 2. This was also found to hold true for

site 2. Consequently, the results lend support for treating team orientation as a

consequence rather than an antecedent of the predictor variables thus rendering it

appropriate to use team orientation as the dependent variable in the analysis.

Having established that is appropriate to use team orientation as the dependent variable,

the next stage is to assess how reasonable the antecedent variables are in predicting

team orientation. This was tested using Ordinary Least Squares (OLS) regression and

involved regressing team orientation on the antecedent and control variables. This

analysis was conducted on the time 2 cross sectional data and the change data. In

82

addition, the analysis was conducted separately on the employee 3 sample from site 1

and site 2. The inclusion of site 2 in the testing of the antecedents of team orientation

will help determine the robustness of the antecedent model. The greater the consistency

of the antecedents in two different organizational contexts, the more robust the model

is. The subsequent section presents and discusses the results of the antecedents of team

orientation using the samples from the two sites.

4.5 An empirical testing of the antecedents of team orientation

Prior to presenting the results, the descriptive statistics are presented for the two sites.

4.5.1 Descriptive Statistics

Table 4.1 presents the results of the t-tests and paired samples t-tests for the individual

site samples. The statistics presented are for the antecedent and dependent variables

used in the model. As mentioned in chapter 3, for ease of clarity, the intercorrelations

between the model variables of each of the results chapters are presented in Appendix

3.

3 Team orientation was not measured at supervisory and managerial levels due to the difficulty in translating the notion of a work group to this level.

83

Table 4.1: Independent t-tests and paired sample t-tests for the individual site samples (employees)

Variables

Site 1 (N=165)

Time 1

Mean (S.D)

Time 2

Mean (S.D)

Change scores

Mean (S.D)

Team orientation 5.73 (0.89)++ 5.60 (0.90)+++ -.13 (.085)* Trust in colleagues 5.67 (0.93)+++ 5.68 (0.91)++ .01 (0.76) Satisfaction with colleagues 5.67 (1.06) 5.61 (0.95)++ -.06 (1.12) Quality awareness 5.99 (0.67)-H-F 5.95 (0.65) -.04 (0.59) -H- Supervisor participative style 4.98 (1.12) 5.15 (1.10)-H-+ .17 (1.08)**-H-F Management commitment to quality 5.10 (1.01)+-1-F 5.27 (1.04)+++ .17 (0.86)**+-H- Improvement in commitment to quality

4.84 (1.06)+++

Site 2 (N=183) Mean (S.D) Mean (S.D) Mean (S.D)

Team orientation 5.50 (1.08)-H- 5.30 (1.15)+++ -.20 (1.16)** Trust in colleagues 5.39 (1.00)+++ 5.42 (1.01)++ .03 (0.99) Satisfaction with colleagues 5.54 (0.91) 5.36 (1.13)++ -.18 (1.24)* Quality awareness 5.79 (0.77)+++ 5.91 (0.69) .12 (0.72)**++ Supervisor participative style 4.85 (1.20) 4.63 (1.34)+++ -.22 (1.41)**+* Management commitment to quality 4.74 (1.17)+++ 4.09 (1.26)+++ -.65 (1.07)***+++ Improvement in commitment to quality

3.98 (1.35)+++

* T-test difference between Time 1 and Time 2 significant at < than .1 level ** T-test difference between Time 1 and Time 2 significant at < than .05 level *** T-test difference between Time 1 and Time 2 significant at < than .01 level + T-test difference in means between Site 1 and Site 2 significant at < than .1 level ++ T-test difference in means between Site 1 and Site 2 significant at < than .05 level +++ T-test difference in means between Site 1 and Site 2 significant at < than .01 level

There has been a significant negative shift in the team orientation at the two sites. This

is the only significant negative change at site 1. Trust in colleagues, satisfaction with

colleagues and quality awareness have remained stable over time for the sample of

employees at site 1. In terms of change, there has been a significant positive shift in

perceived supervisor participative style and perceived management commitment to

quality.

At site 2, in addition to the significant negative shift in team orientation, there have

been significant negative changes in the following: satisfaction with colleagues,

84

perceived supervisor participative style and perceived management commitment to

quality. While trust in colleagues has remained stable over time, quality awareness is

the only variable at site 2 to have significantly shifted in a positive direction.

Comparing the two sites at time 1, significant differences were found in team

orientation, trust in colleagues, quality awareness and perceived management

commitment to quality with site 1 being more positive. At time 2, there are significant

differences in all the variables (except quality awareness) with site 1 being more

positive. In terms of the changes that occurred over time, significant differences were

found between the two sites in changes in quality awareness, supervisor participative

style and management commitment to quality. Regarding the climate measure of

improvement in commitment to quality at the site, site 1 was found to be significantly

more positive.

4.5.2 Results

Table 4.2 presents the results of the impact of the antecedent and control variables on

team orientation using the cross sectional time 2 data for site 1 and site 2. Table 4.3

presents the results of the same analysis using the change data. The standardized beta

coefficients are shown for the antecedent variables. For clarity, the impact of the

control variables are not shown.

85

Table 4.2: The impact of the antecedent and control variables on team orientation at time 2

Antecedents of Team Orientation at time 2 Site 1 Site 2

Satisfaction with colleagues .10 .11+ Trust in colleagues .26*** .24*** Quality awareness .12** .09 Supervisor participative style .22*** .25*** Perceived management commitment to quality .06 .20*** An improvement in commitment to quality at the site

.08 .13**

Adjusted R2 .49 .54 N 165 179

+ = p<.10 ** = p<.05 *** = p<.01

Several points are worth noting. The significant antecedents of team orientation at site

1 come from the individual, colleague and leader categories. However, at site 2, the

significant categories are colleague, leader and climate. The key difference in the

antecedents between the two sites are as follows: quality awareness is significant at site

1 but not at site 2 and perceived management commitment to quality and an

improvement in commitment to quality at the site are significant at site 2 but not at site

1.

86

Table 4.3: The impact of changes in the antecedent variables (and the control variables) on changes in team orientation

Antecedents of changes in team orientation Site 1 Site 2

A in satisfaction with colleagues .21*** .16** A in trust in colleagues .33s** .14+ A in quality awareness .19*** .10

A in supervisory participative style .11 .27*** A in perceived management commitment to quality .05 .10 An improvement in commitment to quality at the site

.07 .08

Adjusted R2 .26 .22 N 164 179

+= p<.10 ** = p<.05 *** = p<.01

In predicting changes in team orientation, there is a greater consistency between the

two sites with the significant antecedents being those that are proximal to the

individual's work environment. At site 1, similar to the cross sectional results, changes

in quality awareness has a significant impact on changes in team orientation.

Similarly, at site 2, a consistent effect on team orientation is supervisor participative

style.

4.6 Discussion

The lack of consistency in antecedents between the two sites is the first issue that needs

to be addressed prior to discussing the results in greater detail. In terns of the

antecedents, the main discrepancies between the two sites centre on the significant

effect of quality awareness at site 1 but not at site 2, the effect of perceived

87

management commitment to quality and an improvement in commitment to quality at

site 2 but absent at site 1. This difference was found in the results of the cross sectional

analysis. Regarding the results of the change data, the difference involves the impact

of changes in quality awareness at site 1 and the impact of changes in supervisor

participative style at site 2. The results raise two questions. First, is it the nature of the

changes that have occurred in the intervening time period at the two sites that has led to

different predictors of changes in team orientation? Second, is it the actual model of

team orientation that is fundamentally flawed?

A weak test of the latter issue would be to investigate the antecedents of team

orientation prior to the changes occurring in the intervening period. This involves

regressing team orientation at time 1 on the antecedent and control variables measured

at time 1.4 The results are presented in Table 4.4 (the control variables are not shown).

As Table 4.4 shows, there is total consistency between the two sites in the antecedents

of team orientation. Overall, the results support the model, in at least predicting

variations in team orientation. This would seem to indicate that the hypothesized

model is not inherently flawed.

4 One of the intervening variables measured at time 2; an improvement in commitment to quality at the site was not measured at time 1. Three other control variables; change in supervisor, change in job content and change of jobs were not measured at time 1.

88

Table 4.4: The impact of the antecedent and control variables on team orientation at time 1

Antecedents of Team Orientation at time 1 Site 1 Site 2

Satisfaction with colleagues .18** .25*** Trust in colleagues .32*** .13+ Quality awareness .21*** .18*** Supervisory participative style .23*** . 33*** Perceived management commitment to quality .12+ .11+

Adjusted R2 .45 .48 N 166 179

+ = p<.10 ** = p<.05 *** = p<.01

Therefore, a more likely explanation for the differential results obtained across the two

sites may be the nature of changes that have occurred at the two sites between time 1

and time 2. This may be a result of different changes triggering different effects.

Broadly speaking, taking the cross sectional and change data, the categories of

colleagues and leader (immediate) seem to be more robust than the individual and

climate categories.

There is consistency between the two sites in the factors that affect changes in team

orientation, these factors are proximal to the individual. This is consistent with the

concept of psychological proximity, from field theory. Consequently, factors closer to

the individual will influence an individual's reaction more immediately than factors

more removed from the individual such as perceived management commitment to

quality and an improvement in commitment to quality at the site. Applying this

argument to change, one would expect changes in the proximal factors to have a more

immediate impact than changes in the more distant factors. The results from the two

sites would provide support for this. This is not to say that the more distant factors are

89

redundant, field theory suggests that the influences of the distant factors are mediated

by the proximal factors. For example, it is quite plausible that management

commitment to quality has an impact on supervisory behaviour which in turn has an

impact on team orientation at the employee level.

An interesting result is that changes in supervisor participative style did not have an

impact on changes in team orientation at site 1. Individuals at site 1 who perceived

their immediate boss as participative reported higher levels of team orientation but that

increased participative behaviour did not lead to higher levels of team orientation.

Overall, at site 1, perceived supervisor participative style significantly increased

between time 1 and time 2. Consequently, a possible explanation for the lack of effect

of changes in participative style on changes in team orientation may be due to the

operation of a threshold effect. What this means is that a certain degree of supervisor

participative style may be a precondition for team orientation but increases beyond a

certain point have no further effect on team orientation.

A crude test of a threshold effect would be to look at the correlation between changes in

supervisor participative style and changes in team orientation for those above and

below the threshold level. If a threshold effect was in operation, one would expect no

relationship between the two variables above the threshold level while a positive one

would exist for those below the threshold level. The sample at site 1 was divided into

two categories at time 1; above and below the threshold level. 5 For individuals above

the threshold level at time 1, the relationship between changes in supervisor

participative style and changes in team orientation was .06 and not significant. For

individuals who were below the threshold level at time 1, the relationship between the

5 The designation of a threshold level seems to be an arbitrary decision. The level chosen in this situation was the mid point where 50% of respondents were below and 50% of respondents were above. This point was 5.2 on a 1-7 scale.

90

two variables was .26 and significant at the .05 level. These results would support the

existence of a threshold effect and also support the contention that beyond a certain

point, greater participation by the supervisor does not result in greater team orientation.

Across the two sites, the categories of colleagues and leader (immediate) seem to be

more robust than the individual and climate categories in predicting variations in and

changes in team orientation. Overall, the significant antecedents of team orientation are

proximal to the individual; colleagues and immediate supervisor. In view of the

different contexts at site 1 and site 2, the model containing the antecedents is broadly

useful as a basis to evaluate the impact of the intervention on team orientation in the

subsequent chapter.

4.7 Conclusions

This chapter examined the antecedents of team orientation. The initial step investigated

what is meant by teamwork. It was argued that two perspectives are evidenced in the

literature; team based working as a structure and team orientation as an individual's

attitude toward his/her work group. Subsequently, an hypothesized model containing

the antecedents of team orientation was presented. This model was empirically tested

using the data from site 1 and site 2.

The results suggest that the significant antecedents consistent between the two sites are

factors that are proximal to the individual. Overall, the model provides a useful basis

for evaluating the impact of the intervention on team orientation. It is to this we now

turn.

91

Chapter 5: The Impact of a TQM Intervention on Team Orientation

5.1 Introduction

The previous chapter presented and empirically tested an hypothesized model of the

antecedents of team orientation. The chapter concluded that the model provided a

useful basis for evaluating the impact of the intervention on team orientation.

Consequently, this chapter sets out to investigate the total impact, if any, of the

intervention on team orientation. Hence, the initial step discusses how the total impact

of the intervention will be assessed; that is, the logic underlying the evaluation.

Reflecting its critical importance, the intervention is examined in detail. This involves

looking at the different aspects of the intervention and their measurement.

Subsequently, the analysis procedures used and the descriptive statistics of the

intervention are discussed. This provides an overview of the extent of participation in

the intervention as well as how the intervention was assessed along a range of

dimensions. Following, the results are presented and discussed.

As discussed in chapter 3, there is a strong argument for assessing the intervention in

terms of its total impact on, in this case, team orientation. This involves assessing the

direct and indirect effects of the intervention on team orientation. The full evaluation

model of the impact of the intervention on team orientation is outlined in Figure 5.1.

There are four intervention variables (these are discussed in greater detail in the

subsequent section) which tap different aspects of the intervention. The first

intervention variables taps employee participation in the intervention. Previous

research has indicated that the assessment of training activities may be more important

92

FIG

UR

E 5.

1: H

ypot

hesi

zed

eval

uatio

n m

odel

of t

he im

pact

of a

TQ

M in

terv

entio

n on

team

ori

enta

tion

MI N■NIL ■•••••1 =I If

Team

ori

enta

tion

N

Inte

r ven

ing

vari

able

s

tz4 cf)

Cd fat

1=1 E

4.)

ci> 4%1

-;1

• no .4.w

(1)

0.) 0

5 0

CCI

, w

e

5 El

cl)

93

than participation in affecting subsequent attitudes (Tannenbaum, Mathieu, Salas and

Cannon-Bowers, 1991). Consequently, the three remaining intervention variables in

the model tap employee assessment of the intervention (perceived benefit and

appropriateness) and their perception of the extent to which their immediate boss is

committed to and involves employees in the intervention.

In the evaluation model, the antecedents of team orientation from the previous chapter

now become the intervening variables. The total impact for each of the four

intervention variables on team orientation is assessed. For example, the direct effect of

participation in the intervention on team orientation is rigorously determined by

controlling for the remaining intervention variables, the intervening variables and a set

of demographic control variables. The indirect effects of participation in the

intervention on team orientation is assessed through its impact on the set of intervening

variables in the model. The direct effect and indirect effects are added to give the total

effect of each of the intervention variables on team orientation.

5.2 Measures of the intervention

As mentioned above, the intervention is operationalized as having four components.

The first component taps an individual's participation in the intervention while the

remaining three components tap different aspects of how individuals assess the

intervention. Several points are worth noting on the measures of the intervention. First,

to my knowledge, no previously validated measures tapping individuals' perceptions

and assessment of a TQM intervention exist. Therefore, the following measures were

created for this study. Second, these measures were collected at time 2. Third, as the

94

measures relate to the TQM intervention, they were collected at site 1 only. Where

appropriate the alpha coefficients for the scales are presented.

Participation in the TQM intervention Respondents at site 1 were asked to what

extent they were participating in the activities of the TQM intervention. Responses

were on a five point Likert scale from not at all to a very great extent. As this is a

single item measure a reliability coefficient was not calculated.

Assessment of the TQM intervention Respondents at site 1 were asked questions

relating to the intervention along the dimensions of perceived benefit of the

intervention, perceived appropriateness of the intervention and supervisory

reinforcement of the intervention.

Perceived benefit of the intervention This four item scale taps respondents'

perception of the extent to which the intervention is of benefit to them; the extent to

which it is part of their job and; the degree to which they think it is a management

initiative to get people to do more work. The alpha coefficient for this scale at time 2

was .75.

Perceived appropriateness of the intervention This four item scale measures the

degree to which respondents felt the intervention was appropriate, and that it was a

priority and of equal benefit to management and employees. The alpha coefficient for

this scale at time 2 was .81.

Reinforcement of the intervention by the immediate supervisor This variable

measures respondents' perception of the degree to which their immediate supervisor

95

was committed to and involved them in the intervention. This three item scale has an

alpha coefficient of .68 at time 2.

5.2.1 Factor analysis of the items measuring the intervention

Factor analysis (principal components method, varimax rotation) was conducted on all

the items measuring perceptions of the intervention. The results indicated a three factor

solution with the items tapping whether the intervention improved management-

employee relations loading on supervisory reinforcement of the intervention. As an

improvement in management-employee relations as a result of the intervention is

essentially an outcome of the intervention (and possibly a consequence of supervisory

reinforcement of the intervention), a second factor analysis was conducted, forcing the

items to load on four factors.' The factors and their loadings are presented in Table 5.1.

1 This step of forcing items to load on a specified number of factors is not uncommon See for example, Scott and Bruce (1994).

96

Table 5.1: Factor analysis of the items measuring the intervention

Item Factor 1 Factor 2 Factor 3 Factor 4

WTTW has resulted in better relations between management and employees .86° .26 .19 .23

WTTW has resulted in greater teamwork between management and employees .82° .31 .21 .22

WTTW has improved communication between management and employees .80° .29 .18 .28

There is a lot of active support among workers in general .67° .06 .32 .27

There is no benefit for me in WTTWI .22 .80° .12 .22

WTTW is not part of my job .07 .79° .03 .19

WTTW is a management initiative to get people to do more workf .19 .64° .27 -.05

WTTW is no better or worse than previous initiativesf .20 .53° .23 -.15

WTTW is an appropriate way to bring about the type of change needed .19 .15 .80° .02

WTTW will benefit me in my job .28 .39 .66° .17

Management and employees will benefit equally from WTTW .17 .41 .58° .24

WTTW is a top priority at this site .23 .19 .53° .50

There is a lot of active support for WTTW among managers f .29 -.05 .48° .47

My immediate boss is strongly committed to WTTW .18 -.03 .26 .79°

My immediate boss involves me in WTTW .20 .34 -.12 .75°

My immediate boss has changed his/her behaviour as a result of WTTW .30 .00 .15 .51°

Eigenvalue 6.6 1.69 1.14 1.00

Percent of variance 42.9 10.6 7.1 6.2

° indicates factor on which item loads most highly f item reversed scored f item dropped as factor loading less than .5

97

The four factors were labelled as follows: improvement in management-employee

relations as a result of the intervention, perceived benefit of the intervention, perceived

appropriateness of the intervention and reinforcement of the intervention by the

supervisor. The first factor cannot be considered as falling into the category of

assessment, it is clearly tapping perceptions linking the intervention to an outcome or

consequence, namely, management-employee relations. 2

Perceived benefit and appropriateness of the intervention are clearly tapping how

individuals cognitively assess the intervention. These represent individuals'

judgements and inherently their expectations of the intervention. The final factor;

supervisory reinforcement of the intervention does not directly tap how individuals

assess the intervention. Rather, it taps an individual's assessment of the process of

implementation or the cascading of the intervention. In this sense, it taps an assessment

of the process rather than an assessment of the intervention per se.


The evaluation model was tested in two stages using the data from site 1. The first

stage was to regress the intervening ( antecedents from the previous chapter) variables

on the intervention variables and control variables at site 1. The second stage was to

regress the dependent variable, team orientation, on all the variables in the model.

Separate regressions were carried out to obtain the path coefficients (standardized beta)

for the two stages of the model. The path coefficients were used to measure the direct,

indirect and total effects of the variables in the model. The regressions were carried out

2 Therefore, it is excluded from the analysis in this and subsequent chapters. However, as it is an outcome of the intervention, it is picked up in Chapter 11 as part of an overall assessment of the impact of the intervention.

98

on the time 2 and the change data. As the change data is arguably the more rigorous,

greater emphasis will be placed on these results in the discussion.

5.4 Descriptive statistics of the TQM intervention

This section looks in detail at the intervention variables in view of their centrality to the

evaluation. As a starting point, the means and standard deviations are presented for the

four intervention variables in Table 5.2.

Table 5.2: Descriptive statistics of the TQM intervention variables

Time 2 Mean (S.D)

Participation in the intervention 2.58 (1.15) Perceived benefit of the intervention 4.20 (1.31) Perceived appropriateness of the intervention 4.99 (1.26) Reinforcement of the intervention 4.56 (1.28)

First, participation in the intervention was measured as a continuous variable so that

different degrees of participation could be captured. This also provided a way to reflect

the complexities of the intervention and was deemed more appropriate than asking

respondents if they were participating in the intervention eliciting a 'yes/no' response.

To measure the extent of participation, a five point Likert scale was used ranging from

not at all to a very great extent.

Individuals who responded in the 'not at all' or 'not much' category were aware of the

intervention and had received communication about the intervention when it was

99

launched. In essence, these individuals were not participating in the intervention. 78

out of 165 employees fell into these two categories. In contrast, individuals who

responded in the "to a very great extent" and "to a great extent" had received training

by their supervisor and were participating in teams with the aim of making

improvements in their work area. 32 out of 165 fell into these two categories. The

remaining 55 employees responded in the "to some extent" category. These

individuals were not participating in teams. Their involvement in the intervention was

limited to either being trained in the principles of the intervention by their supervisor or

being involved at a more informal level by their supervisor. This could mean that the

supervisor has communicated information about the intervention but has not given their

employees training or that the employee was doing intervention related activities to

assist the supervisor in his/her cross functional or hierarchical team.

In terms of perceived benefit of the intervention, 34.6% of employees thought the

intervention was beneficial ( mean response of 5.0 or above), 43% of employees did not

(to varying degrees) think that the intervention was beneficial ( mean response of 3.75

or below) and 22.4% remained relatively neutral on the benefit of the intervention

(mean response greater than 3.75 and less than 5.0). A more positive picture emerges

from employees' views of the appropriateness of the intervention; 50.9% thought it was

appropriate to varying extents ( mean response of 5.0 or above), 18.8% did not perceive

the intervention as appropriate (3.75 or below) and 30.3% remained relatively neutral

on the appropriateness of the intervention (mean response greater than 3.75 and less

than 5.0). 43% of employees perceived their supervisor as reinforcing the intervention

(mean response of 5.0 or above), 37% did not think their supervisor was reinforcing the

intervention (mean response of 4.0 or below) while the remaining 20% did not have a

clear view of their supervisor reinforcing the intervention (mean response greater than

4.0 and less than 5.0).

100

Overall, a varied picture is presented of the intervention in terms of employee

participation and assessment. A balanced variation is found regarding employee

participation with approximately half of the employees participating to varying extents

while the remaining employees are not participating. In terms of employee assessment

of the intervention, there is considerable variation between employees. Approximately

half of the employees viewed the intervention as appropriate while a third of employees

judged the intervention to be beneficial. Just over 40% of employees perceived their

supervisor as reinforcing the intervention.

5.5 Results

This section presents the results of the evaluation model of team orientation. Table 5.3

presents the direct, indirect and total effects of the intervention variables on team

orientation at time 2.

101

Table 5.3: Direct, indirect and total effects of the TQM intervention on variables in the model at Time 2

TQM Variables

Sat. with Coll.

Trust in

Coll. Quality aware

Sup. partic. style

Mgt. commit

to quality

Improve in

commit. to quality

Team orientation

1. Participation in TQM Direct Effects on -.06 -.02 .05 -.01 .05 .04 .13+

Indirect Effects on -.01 Total Effect .12

2. Perceived benefit Direct Effects on .00 .00 -.04 .19** .14** .24** -.07 Indirect Effect on .04

Total Effect -.03

3. Perceived appropriateness

Direct Effects on .11 -.05 .04 -.05 .08 .19** .09 Indirect Effect on .00

Total Effect on .09

4. Supervisory reinforcement

Direct Effects on .16+ .30** .07 .29*** .15** .21** .09 Indirect Effect on .16

Total Effect .25***

Adjusted R2 .17 .49 .36 .39 .51 .29 .48 N 165 165 165 165 165 165 165

+ = p<.10 p‹.05 *** = p<.01

The results show that participation in the TQM intervention did not have any significant

effect on the intervening variables. Furthermore, participation did not have a

significant effect on team orientation. Second, it is the assessment of the intervention

rather than participation per se that is important in affecting the intervening variables.

In particular, supervisory reinforcement of the intervention had a stronger effect on the

intervening variables than did the other two assessment variables. Finally, one aspect

of the intervention, supervisory reinforcement had a significant total effect on team

orientation. In brief, the linkage found between the TQM intervention and team

102

orientation is an indirect one. It occurs through supervisory reinforcement affecting

trust in colleagues and supervisor participative style which in turn affect team

orientation.

Table 5.4 presents the direct, indirect and total effects of the TQM variables on changes

in team orientation between the two occasions of measurement.

Table 5.4: Direct, indirect and total effects of TQM intervention on variables in the model -change over time (T2-T1)

TQM Variables

Sat. with Coll.

Trust in

Coll. Quality aware

Sup. partic. style

Mgt. commit

to quality

Improve in

commit. to quality

Team orientation

1. Participation in TQM Direct Effects on -.10 -.05 .11 -.11 .15 .04 .07

Indirect Effects on -.02 Total Effect .05

2. Perceived benefit Direct Effects on .01 -.05 -.09 .19+ .11 .24*** .04 Indirect Effect on -.01

Total Effect .03

3. Perceived appropriateness

Direct Effects on .10 -.03 -.06 -.17 -.03 .19** -.04 Indirect Effect on -.01

Total Effect on -.05

4. Supervisory reinforcement

Direct Effects on .06 .26** .05 .14 .09 .21** .09 Indirect Effect on .13

Total Effect .22***

Adjusted R2 .00 .00 .00 .04 .05 .29 .25 N 165 165 165 165 165 165 165

+ = p<.10 ** = p<.05 *** = p<.01

103

The results of the change analysis confirm the cross sectional results. Participation did

not have a significant impact on changes in the intervening or dependent variables. The

results confinn the importance of assessment of the intervention over participation in

affecting subsequent attitudes. Of the assessment variables, supervisory reinforcement

had a significant effect on team orientation. Finally, the same linking mechanism was

found between the TQM intervention and changes in team orientation.

5.6 Discussion

The impact of the intervention will be discussed in two stages beginning with its impact

on the intervening variables and then looking at its impact on the dependent variable.

Intervening variables

Overall, the intervention had some impact on the intervening variables. Of the

intervention variables, supervisory reinforcement had the greatest impact. In particular,

supervisory reinforcement was found to have a significant effect on variations and

changes in trust in colleagues (which subsequently affected team orientation). Previous

research has identified three classes of determinants of trust; personality characteristics,

interpersonal characteristics and situational features (Golembiewski and McConkie,

1975). It is the latter class of determinant that may help explain why supervisory

reinforcement of the intervention would have an effect on trust in colleagues.

Golembiewski and McConkie (1975) present three methods which seek to increase trust

in organizations; group decision making to reduce mistrust; reorganization of work to

encourage integration and collaboration and; efforts to change norms toward greater

openness and problem solving. These methods broadly reflect what TQM purports to

achieve. The TQM intervention at the work group level is directed toward uniting the

104

group toward problem solving and collaborative efforts to improve quality. For this to

be accomplished, it is necessary for the immediate supervisor to be committed to the

intervention and transfer the intervention to the workplace by involving group

members. Allowing group members to become involved in the intervention will

provide a legitimate forum for greater group interaction, facilitate greater awareness of

each individual's role in the group and provide a superordinate goal of continuous

improvement that requires collaboration among members. As a result, one would

expect positive changes in trust in colleagues.

Two aspects of the intervention have a significant impact on supervisor participative

style. First, supervisory reinforcement has a significant positive effect. The

explanation is straightforward as supervisors who are reinforcing the intervention are

involving subordinates in the intervention and thus facilitating their participation.

Consequently, it is not surprising that individuals who perceive their immediate boss as

reinforcing the intervention are more likely to see their immediate boss as participative.

Second, individuals who perceive the intervention as providing some benefit to them

are more likely to perceive their supervisor as participative. As mentioned earlier, an

assumption underlying this chapter is that the four TQM intervention variables are

exogenous in relation to the other variables in the model. In this case, however, the

direction of causality is less clear; that is, supervisor participative style prior to the

intervention may be a significant factor in influencing an individual's assessment of the

intervention. Taking this further, it is quite likely that participatively oriented

supervisors prior to the intervention are more likely to communicate information

regarding the intervention to subordinates which consequently shape their assessment

of the intervention. The importance of supervisors as providers of social information

has previously been highlighted (Griffin, 1983). Thus, rather than perceived benefit

105

affecting supervisor participative style, the direction of influence may be in the contrary

direction. This issue is pursued in greater detail in chapters 8 and 9.

The impact of supervisory reinforcement on perceived management commitment to

quality highlights the saliency of the immediate boss. Leadership behaviours are likely

to be interpreted as representative of the wider organization (Kozlowski and Doherty,

1989) and thus individuals are likely to attribute supervisory behaviour to management.

This linkage may be reciprocal as management actions and policies are likely to have

an effect on supervisory behaviour. Thus, supervisors reflect management actions,

policies and procedures.

The intervention had no significant impact on one of the intervening variables, quality

awareness. One explanation for the lack of change in quality awareness over time may

be the use of an insensitive instrument. The quality awareness scale yielded responses

toward the top end of the scale indicating the potential operation of social desirability.

The distribution and standard deviation (around 0.65) would support this. Differences

in quality awareness across job titles provides counter support to the insensitivity

argument. A more plausible explanation is that the intervention did not provide a

convincing rationale that there was a real need to improve the awareness of one's own

actions and their consequences.

Team orientation

Turning now to the impact of the TQM intervention on team orientation, two main

findings stand out. First, the intervention did not have a significant direct impact on

team orientation. Thus, none of the intervention variables had a significant direct effect

on changes or variations in team orientation. However, the intervention did have a

significant total effect on team orientation. In particular, supervisory reinforcement of

106

the intervention had a significant impact on an individual's trust in their colleagues

which had a significant effect on team orientation. The nature of these linkages have

previously been discussed.

Second, and more generally, the results suggest that the assessment of the intervention

is a better predictor of changes in the intervening and dependent variables than

participation in the intervention per se which was not found to have a significant effect

on any of the variables in the model. The importance of individual reactions to the

intervention is consistent with previous research on training which suggests that

participation in training activities is not sufficient to bring about attitudinal and

behavioural change Rather, it is training fulfilment that played a central role in the

development of posttraining attitudes. (Tannenbaum, Mathieu, Salas and Cannon-

Bowers, 1991).

It is also in line with research on Quality Circles (QCs), which hints that employee

reactions to QC activities may play a central role in explaining the development of

subsequent attitudes. For example, Griffin (1988) found that positive effects of

participation in QCs (job satisfaction and organizational commitment) were evidenced

in the initial 18 month period of QC operation. However, these positive effects

subsequently declined to the baseline level. While the author did not measure

employee reactions to or assessment of QCs, he offers some possible explanations for

the decline in positive effects, one of which is the perceived decline in management

support by QC participants. Bruning and Liverpool (1993) found that participation in

QCs was not a significant predictor of outcome variables; respondents failed to show

more positive outcomes if they were members of QCs. Similarly, the authors did not

include measures tapping employee evaluation of QCs. They, however, suggest that

management attitudes toward QCs could affect an individual's experience of QC

107

activities. Thus, for example, participants' assessment of the support received from

management for QCs may be an important factor in explaining the lack of predictive

power of participation per se .

Most notably, the results presented here stress the importance of reactions to the

intervention in predicting subsequent attitudes. How employees react and what predicts

how they react is investigated in a subsequent chapter. If these findings are broadly

correct, they have important implications for evaluating employee participation

programs

Participation in the intervention

The lack of predictive power of participation in the intervention on subsequent attitudes

is interesting in view of the considerable emphasis placed on participation per se in

TQM and QC research. A number of explanations are offered for this finding. First, it

is possible that participation in the intervention may not have matched pre participation

expectations thus having no effect or a damping effect on subsequent attitudes. This is

partly conjectural as the questionnaire did nor include items tapping met expectations

but from interviews at the site it was noted that the intervention was launched in quite a

dramatic fashion but it took considerable time for the training to be completed and the

mechanisms to be set up. This would lend support to high expectations being created at

the launch of the programme which were not met during the initial period.

Second, much of the literature on Total Quality and Quality Circles stresses the

voluntary nature of employee participation. In practice, little is known about the extent

to which individuals are put under pressure from peers and/or their immediate boss to

participate. The extent to which an individual feels under pressure to participate may

explain the lack of change in subsequent attitudes. A final point that may shed light on

108

the poor predictive effect of participation has to do with the time span of the study. It

may be the case that the lapse of time between an individual's participation and the

collection of attitudinal data was not long enough to reasonably expect participation to

have an impact. Griffin (1988) in his longitudinal study of QCs allowed six months to

lapse between the baseline questionnaire and the first questionnaire after the

commencement of QCs. This time span is not uncommon in the initial stage of the

evaluation of change (see Wall et al. (1986) for an initial measurement 6 months after

the commencement of AWGs). It may be the case that in the longer term as the

intervention develops and takes root, participation may have a greater predictive power

than the assessment of the intervention. Further research is needed to assess the impact

of different time lags on the predictive power of participation in TQM interventions on

subsequent attitudes

Decrease in team orientation

So far, the discussion has focused on how the TQM intervention has had an impact on

team orientation. The intervention had a significant total effect on team orientation.

The final issue that needs to be addressed is why the mean score of team orientation for

the employee sample as a whole has decreased (10% significance level) between time 1

and time 2.

A possible explanation may be that other changes may have affected team orientation in

a negative way. The major change that occurred at site 1 was the TQM intervention.

From my knowledge, there were no changes within the site that would have had a

negative impact on team orientation. Second, at the level of the work group, one may

argue that movement of individuals between different work groups may have affected

team orientation. This may be eliminated as an explanation as there was a negligible

movement of individuals between different work areas so individuals would have the

109

same colleagues at time 1 and time 2. It may be worthwhile reverting attention to the

TQM intervention in order to explain why a decrease in team orientation occurred in

the intervening period.

In pursuing this possible route, it may be beneficial to assess whether a particular group

of employees can be identified whose team orientation shifted in a negative direction.

To explore this further, the sample was divided into two subgroups; participants and

non participants in the TQM intervention. Paired sample t-tests were conducted for the

two groups on team orientation. The results showed that over time, team orientation

remained unchanged for participants (5.93 at time 1 and 5.90 at time 2). What is

interesting is that team orientation has shifted in a negative direction for non

participants (5.51 at time 1 and 5.28 at time 2, significant at the 5% level). While these

results are important in attempting to identify why team orientation has decreased, they

do not indicate that participation is an important predictor of team orientation as no

other variables are controlled for. What the results allow is a greater accuracy in

pinpointing the group of individuals for which team orientation has decreased. Why

would team orientation decrease for non participants?

It would seem that a negative reaction has occurred in the group of individuals who

have not participated in the TQM intervention. This may be a consequence of raised

expectations followed by disillusionment when expectations are not met. Bennis

(1977) argues if changes interventions are badly managed, they may have negative

rather than the desired positive consequences. There are implications for managing

TQM change interventions. From an implementation stance, it may be important to

plan a rapid diffusion of the program throughout the organization. This would be

consistent with the recommendation of Graham and Verma (1991) that there is a need

to manage "proximity" to ensure positive affect toward employee participation

110

programs. The decrease in team orientation may be a short term effect reflecting a

feeling of being "left out" of the intervention which will dissipate once the intervention

has diffused more widely.

To the extent that one can argue that non participants have reacted in a negative

manner, this points to a diffusion problem. However, a more serious issue is that

participants have not significantly changed their team orientation. In order to put

forward a possible explanation for this finding, it is necessary to look at the perceptions

that participants hold regarding their immediate supervisor's reinforcement of the

intervention. Thus, if supervisors diffuse the intervention (involving subordinates in

the intervention) and are visibly committed to it, one would expect that participants

would report their supervisors as reinforcing the intervention. Looking at the group of

participants, 51.6% report that their supervisor is reinforcing the intervention (mean

score of 5.0 or above) while 17.2% report that their supervisor is not reinforcing the

intervention (mean response of 4.0 or below). The remaining 31.2% remain relatively

neutral on their supervisor reinforcing the intervention. 3

What this may suggest is that approximately half of the participants perceive their

supervisor as committed to and supportive of the intervention. Consequently, employee

participation in the intervention does not necessarily mean that their supervisor is

committed to the intervention. In essence, a fine distinction is necessary between what

may be supervisor compliance (involving subordinates in the intervention) and

supervisor commitment (not only involving subordinates but also supporting and

reinforcing the intervention). The fact that nearly half the participants do not report

their supervisor as positively reinforcing the intervention may help explain why team

orientation has not significantly changed in a positive direction.

3 For non participants, 17.7% report their supervisor as reinforcing the intervention while 59% report their supervisor as not reinforcing the intervention.

111

5.7 Conclusions

This chapter assessed the impact of a TQM intervention on team orientation. The

impact of the TQM intervention on team orientation was found to be an indirect one.

The linkage occurred through supervisory reinforcement of the intervention which

affected trust in colleagues which in turn affected team orientation.

The results highlighted that participation in the intervention per se was not a sufficient

condition to affect subsequent attitudes. Rather, the significant aspect of the

intervention was employee perception of the extent to which their immediate boss

reinforced the intervention. However, team orientation shifted significantly in a

negative direction for non participants and remained unchanged for participants

between time 1 and time 2. A potential explanation put forward was the lack of

supervisory reinforcement in terms of involving employees which resulted in a feeling

of being 'left out' by non participants. Regarding participants, even though they are

participating in the intervention, nearly half did not perceive their supervisor as

positively reinforcing the intervention. This would seem to indicate that employee

participation in the intervention does not automatically mean that the supervisor is

reinforcing the intervention. A supervisor may comply by involving subordinates in the

intervention but this does not mean that the supervisor supports and reinforces the

intervention.

Does the intervention have a similar impact on the other key expected outcomes of

TQM? The next two chapters address this question in two stages. The first stage

examines the concept of commitment to improvement itself and assesses its similarity

to intrinsic motivation. The second stage (chapter 7) involves assessing the impact of

the intervention on these concepts.

112

Chapter 6: Commitment to Improvement and Intrinsic Motivation: The Same Constructs?

6.1 Introduction

The previous chapter investigated the impact of a TQM intervention on team

orientation. The results showed that the impact was indirect; that is, supervisory

reinforcement of the intervention had a significant effect on trust in colleagues, which

in turn, had a significant influence on team orientation. This raises the question of

whether the effect of the intervention on team orientation is mirrored for other core

elements of TQM. Therefore, this line of investigation continues and assesses the

impact of the intervention on commitment to improvement and intrinsic motivation.

Several concepts such as commitment to quality, commitment to continuous

improvement and intrinsic motivation are widely used in the literature in what seems to

be an interchangeable manner On one hand, Deming (1986) Crosby (1986) and Dale

and Cooper (1992) place great emphasis on motivation. They argue that employee

demotivation is a result of primarily management practices. The prescription is quite

clear; by changing management practices through adopting and pursuing TQM,

employees will experience greater motivation. Here, the underlying assumption is that

employees are in fact motivated and want to do a good job, if this does not occur, there

is a strong likelihood that the reason is due to management practices. Juran (1989) and

Wilkinson (1992) argue that by giving employees responsibility for the quality of their

work, this will improve their motivation. Feigenbaum (1983) uses the term

commitment to quality to capture what seems to be an orientation to doing a quality

job.

113

More recently, the term empowerment has entered into the TQM arena (Grant et al.,

1994; Cruise 0' Brien, 1995). The term has its origins outside TQM (Bennis and

Nanus, 1985; Burke, 1986) and has been interpreted in a variety of ways, for example,

to describe specific interventions as well as the presumed effects of these interventions

on employees (Conger and Kanungo, 1988). It has been recommended that

empowerment be defined in terms of motivational processes; that is, an increase in

workers' effort - performance expectancies (Conger and Kanungo, 1988, p4-75) or

increased task motivation (Thomas and Velthouse, 1990). In this light, TQM

interventions may be viewed as empowering through managerial actions to remove

obstacles that muddy the link between effort and performance thereby increasing

intrinsic motivation. In this sense, empowering is the act of taking away demotivators

(Perisco, 1991).

In addition some writers emphasize continuous improvement rather than motivation.

Hill (1991b) states that one element of a quality culture is the 'internalization of quality

and continuous improvement as the goal of all activities' (p555). In more recent

contributions (Waldman, 1994; Sitkin et al, 1994), the importance of continuous

improvement as a key outcome of TQM is reinforced. Emphasizing this perspective,

the notion of continuous improvement is akin to innovation; problem recognition,

generation of new ideas or solutions, sponsorship of the new idea and the

institutionalization of the new idea (Scott and Bruce, 1994; Kanter, 1988). Lawler

(1994) hints at the distinctiveness of intrinsic motivation and continuous improvement.

He argues that employees have responsibility in two areas: calling attention to quality

problems as they do their normal work and accepting the continuous improvement

culture by looking for ways to improve how they do their work and ways to improve

the overall activities of the organization. The former area of responsibility can be

thought of as intrinsic motivation; that is, the desire to do a good job and the

114

notification to management of barriers to accomplishing this. The latter may be

thought of as commitment to improvement.

In view of the parallel usage of intrinsic motivation and commitment to continuous

improvement, the distinction (or lack of) between the concepts warrants investigation.

Prior to assessing the impact of the intervention on commitment to improvement and

intrinsic motivation, it is necessary to assess whether these two concepts are indeed

conceptually similar and thus interchangeable. Consequently, the aim of this chapter is

to examine the two concepts thus providing the basis for investigating the impact of the

intervention on intrinsic motivation and commitment to improvement in the subsequent

chapter. In addition, a secondary aim of the chapter is to present and test a model of

commitment to improvement.

To my knowledge, there has been little attempt to operationalize commitment to

improvement. The chapter begins by proposing a definition of commitment to

improvement. This provides a starting point for examining its similarity with intrinsic

motivation. Factor analysis is used to examine the nature of the underlying constructs

being measured by the two concepts. A model of commitment to improvement is then

presented. The second stage of the investigation uses the model to assess the degree of

commonality between the two concepts in relation to their predictors.

6.2 Commitment to improvement

This section discusses the nature of commitment to improvement and a definition is

proposed. Following this, the items measuring commitment to improvement are factor

analysed to assess the dimensionality of the concept. This provides the basis for the

115

initial assessment of the conceptual similarity of commitment to improvement and

intrinsic motivation.

The concept of continuous improvement is a key component of TQM. In the context of

TQM, work performance defined in terms of in-role behaviour (Campbell, 1990) may

not be wholly appropriate (Waldman, 1994) to a TQM culture. Individuals are

expected to accomplish their formally defined tasks, directing attention to obstacles that

prevent such accomplishment. They are also required to think about ways of how they

can improve the way they do their job, the work of their group and overall activities of

the organization. Thus, commitment to improvement goes beyond an individual doing

the best job they can, it requires an individual to generate ideas for improvement and

take action on those ideas.

6.2.1 Definition of commitment to improvement

Commitment to improvement is defined as an individual's internalization of continuous

improvement and a willingness to exert effort to find ways to prevent mistakes and

make improvements. It is viewed not only in terms of attitude but also in terms of

behaviour. In addition, commitment to improvement is hypothesized to have two

components; general orientation to quality and improvement as part of the job. This

definition does not include an individual's receptivity to improvement ideas coming

from others which arguably may be another dimension of commitment to improvement.

The first component of commitment to improvement is labelled a general orientation to

quality and was measured by four items in the questionnaire:

- In my work area, I am always looking for ways to prevent mistakes

116

- I have put a lot of effort into thinking about how I can improve my work

- To know I had made a contribution to improving things around here would

please me

- I am strongly committed to Total Quality

The first two items reflect the behavioural dimension; that is, an individual's

willingness to look for ways to prevent mistakes in the work area and an individual's

willingness to exert effort in looking for ways to improve his/her job. The latter two

elements represent a general affective orientation to quality and improvement.

The second component is labelled improvement as part of the job. This reflects a

greater internalization of improvement; that is, an individual feels that looking for

improvements is an inherent part of their job. It was measured with the following three

items:

- Looking for ways of improving how things are done around here is part of

my job

- I am not paid to think of ways of improving things around here

- I often put forward ideas and suggestions without expecting extra reward

The first two items tap an affective dimension, the extent to which improvement is

considered part of the job and the extent to which individuals think they are paid to

think of ways to contribute to improvements. The last item taps behaviour, the extent

to which an individual puts forward ideas and suggestions without expecting extra

reward.

117

6.2.2 Factor analysis of commitment to improvement

Factor analysis (principal components method, varimax rotation) was conducted on the

seven items measuring commitment to improvement at time 2 using the combined site

1 and site 2 sample. The results revealed two factors which may be thought to relate to

degrees of commitment to improvement and reflect the two components identified

above; general orientation to quality and improvement as part of the job. These factors

are displayed in Table 6.1.

Table 6.1: Factor analysis of items measuring commitment to improvement

Item

Factor 1 Factor 2

I am strongly committed to Total Quality .80° .02

To know I had made a contribution to improving things around here would please me .73° .21

In my work area I am always looking for ways to prevent mistakes .68° .21

I have put a lot of effort into thinking about how I can improve my work .60° .42

Looking for ways to improve how things are done is part of my job .22 .78°

I am not paid to think of ways of improving things j' .12 .72°

I often put forward ideas and suggestions without expecting extra reward .14 .67°

Eigenvalue 2.88 1.02 Percent of variance 41.2 14.7

° indicates factor on which item loads most highly t item reversed scored

118

The results of the factor analysis were replicated using the data from the individual site

samples (Appendix 10 and 11). This provides support for treating commitment to

improvement not as a unidimensional concept but as having two distinct dimensions.

The two dimensions may be thought of as levels of commitment to improvement;

general orientation to quality being easier to affect than improvement as part of an

individual's job. In view of these two dimensions, the model outlined later will be

tested separately on general orientation to quality and improvement as part of the job.

6.3 Intrinsic motivation: A different construct?

Having established that commitment to improvement is comprised of a general

orientation to quality and improvement as part of the job, the next step is to examine at

a conceptual level potential differences between these two concepts and the notion of

intrinsic motivation. One could argue that all three concepts are measuring an

underlying desire to perform well; that is, an individual's desire to do the best possible

job he/she can. However, it is plausible that the three concepts are tapping related but

distinct constructs. For example, intrinsic motivation taps the degree to which an

individual wants to work well in his/her job in order to achieve intrinsic satisfaction. In

contrast, improvement as part of the job focuses on the degree to which an individual

feels that looking for improvements is an integral part of their job. Consequently, it

could be argued that both of these concepts are tapping particular orientations toward

the job. This is where the similarity ends; intrinsic motivation focuses on performance

to achieve intrinsic satisfaction. Improvement as part of the job differs from intrinsic

motivation in two important respects. First, it focuses exclusively on the degree to

which an individual feels that looking for improvements is an integral part of their job

and second, it does not propose that an individual wants to make improvements in order

to experience intrinsic satisfaction.

119

A general orientation toward quality may be more conceptually similar to intrinsic

motivation than improvement as part of the job. However, intrinsic motivation clearly

focuses on the job the individual is doing, general orientation to quality is not job

specific. It taps an individual's willingness to prevent mistakes and contribute to

improvements not just in their job but in their work area as well.

The possibility that commitment to improvement and intrinsic motivation are

conceptually distinct has been raised. Consequently, the next step would be to test this

proposition empirically using the data from the two sites. To do this, the items

measuring commitment to improvement and intrinsic motivation were factor analysed

together. Intrinsic motivation was measured using Warr, Cook and Wall's (1979) six

item scale. The results of the factor analysis using principal components and varimax

rotation for the combined site 1 and site 2 sample are shown in Table 6.2.

120

Table 6.2: Factor analysis of items measuring intrinsic motivation and commitment to improvement

Item

Factor 1 Factor 2

I take pride in doing my job as well as I can .80° .12

I feel unhappy when my work is not up to my usual standard .79° .08

I feel a sense of personal satisfaction when I do this job well .77° .21

I try to think of ways of doing my job effectively .74° .27

I like to look back on the day's work with a sense of a job well done .72° .23

My opinion of myself goes down when I do this job badly .64° -.05

To know I had made a contribution to improving things around would please me 47° .47

I am strongly committed to Total Quality 45° .38


I am not paid to think of ways of improving thingst -.02 .70°


Put a lot of effort into thinking about how I can improve my work .37 .59°

Always looking for ways to prevent mistakes .41 .45°

Eigenvalue 5.13 1.60

Percent of variance 39.5 12.3

indicates factor on which item loads most highly t item reversed scored

121

The results indicate the presence of two factors that broadly correspond to commitment

to improvement and intrinsic motivation. However, two items of general orientation to

quality have relatively similar loadings on factor 1 and 2. If one looks at the results of

the factor analysis for the individual site samples (appendix 12 and 13), then intrinsic

motivation and commitment to improvement were found to be factorially independent.

However, commitment to improvement was not found to have two factors

corresponding to the components outlined previously. This questions the factorial

independence of general orientation to quality and improvement as part of the job. In

view of the aim of this chapter; to assess the conceptual similarity of commitment to

improvement and intrinsic motivation, it is worthwhile treating the two components of

commitment to improvement as independent so that any conceptual similarity found to

intrinsic motivation can be more accurately attributed to one or the other dimension of


Overall, the results of the factor analysis broadly support commitment to improvement

as being a different concept from intrinsic motivation. The next step involves

comparing the antecedents of general orientation to quality, improvement as part of the

job and intrinsic motivation. In order to do this, a general model of commitment to

improvement is first presented. This model is then tested in relation to the three

concepts separately. This will allow a comparative assessment of the degree of

similarity in the antecedents of the three concepts.

122

6.4 The antecedents of commitment to improvement

The hypothesized model containing the antecedents of commitment to improvement is

outlined in Figure 6.1. Commitment to improvement is viewed as an outcome of three

systems; individual, leader and organizational. At the individual level, attitudes toward

quality, work and the organization are hypothesized to positively affect an individual's

commitment to improvement. The antecedents in the leader category include the

attitudes and behaviour of the immediate supervisor as well as that of management

which are hypothesized to positively impact an individual's commitment to

improvement. Finally, the climate of the organization is hypothesized to affect an

individual's commitment to improvement. The predictors are now discussed in greater

detail.

Individual level factors. There are three individual level factors which are

hypothesized to have an impact on commitment to improvement. These are quality

awareness, higher order need strength and organizational commitment. Quality

awareness, which as we saw in chapter 4, refers to individuals' awareness of the

consequences of their own actions and the importance they assign to their own work

and to continuous improvement for the success of the organization, is hypothesized to

have a positive impact on commitment to improvement. The rationale is similar to that

of task significance in work redesign (Hackman and Oldham, 1980). Task significance

is the extent to which a job has impact on people in the immediate organization or in

the world at large (p78-79). In a TQM setting, this would involve an awareness that

one's own actions (i.e. the quality of work) will have implications for others. Hackman

and Oldham (1980) hypothesize that task significance, in addition to two other job

characteristics, will lead to high internal work motivation.

123

NNN

k

Com

mitm

ent t

o im

prov

emen

t

124

Most of the research on higher order need strength has been based on the nature of its

effect, that of moderating the relationship between the design of work and outcomes

such as motivation and satisfaction (Berlinger, Glick, and Rodgers, 1988; Hackman and

Oldham, 1980; Spector, 1985). The underlying premise is that individuals with high

levels of higher order need strength are more likely to experience increased motivation,

satisfaction and performance from jobs high in scope. In a similar vein, Rafaeli (1985)

suggests that growth need strength may moderate employees' reactions to quality circle

activities. There is mixed evidence for higher order need strength as a moderator

(Spector, 1985). In the present model, higher order need strength is treated as a positive

predictor of commitment to improvement. In other words, individuals who have a high

need for achievement and satisfaction through work are hypothesized to be more likely

to be committed to improvement. This is based on the assumption that engaging in

continuous improvement is a source of satisfaction and achievement for individuals.

There is a growing body of literature that suggests that organizational commitment per

se is a desirable outcome. Taking this further, one could argue that individuals who are

strongly committed to the organization are more likely to accept the goals of the

organization and exert effort to attain those goals. Thus, if the organization has as its

goal continuous improvement, highly committed individuals are more likely to be

committed to continuous improvement. There is some evidence that committed

individuals are more likely to engage in extra role behaviours such as creativity and

innovation (Katz and Kahn, 1978). Kanter (1968) suggests that commitment translates

into greater efforts by workers. In a similar vein, Eisenberger, Fasolo and Davis-

LaMastro (1990) found that perceived organizational support from the organization was

positively related to organizational commitment and employee innovation (suggestions

for improving organizational operations). The authors argue that the results suggest

125

that citizenship behaviour is a consequence, partly, of an affective attachment to the

organization which is consistent with previous research (Brief and Motowidlo, 1986;

Organ, 1988). In line with the above arguments, organizational commitment is

hypothesized to have a positive impact on individuals' commitment to improvement.

Leader level factors. The next two predictors relate to the attitudes and behaviour of

leaders; management and the supervisor. In the TQM literature, great emphasis is

placed on management commitment to quality as the mechanism for instilling quality

values and behaviours throughout the organization (Deming, 1986; Feigenbaum, 1983;

Hill, 1991b; Waldman, 1994). In Crosby's (1986) view, ".... the attitude of employees is

a clear result of what they see in the attitude of senior management" (p. 196).

Similarly, Dale and Cooper (1992) argue that senior management is responsible for the

"organizational culture, behaviour, values, climate and style of management in which

TQM will either flourish or wither" (p. 43). Cultural change is the objective of TQM

(Hill, 1991b; Wilkinson, 1994) and management has a key role to play in affecting

culture change (Schein, 1985). One mechanism for transmitting culture is the actions

of leaders, "what leaders pay attention to, measure and control" (p.225). Thus, the

greater the perceived management emphasis on quality and continuous improvement,

the greater the commitment to quality and continuous improvement amongst

organizational members.

Supervisory commitment to quality is included in this model in addition to

management commitment to quality for two reasons. First, previous research has

argued that supervisors are the most salient representatives of management actions,

policies and procedures (Kozlowski and Doherty, 1989). Second, the concept of

psychological proximity, from field theory (Lewin, 1943) is relevant. The basic premise

is that factors more proximal in the work environment should influence an individual's

126

reactions more immediately than factors more removed from the individual. Scott and

Bruce (1994) found that supervisory expectations regarding subordinate innovation

were positively correlated to subordinates' innovative behaviour. The authors argue

that this was due to the presence of the "Pygmalion effect" (Livingston, 1969). The

Pygmalion effect refers to the modification of a focal individual's behaviour based on

the expectations for that behaviour received from another (Eden, 1984). Expectations

of subordinate behaviour are communicated to them through supervisory behaviour.

Consequently, the quality oriented behaviour of the supervisor will shape the quality

behaviour of subordinates. Therefore, these perspectives would suggest that the

immediate boss will be more salient than management as a group in influencing

subordinate commitment to improvement.

Third, the focus of TQM writings on senior leadership implies that transactional

leadership theory which emphasizes the work group level, is unimportant to TQM

(Dean and Bowen, 1994). Within the context of TQM, Dean and Bowen (1994) argue

that far less importance is attached to the role of leadership further down the hierarchy.

Therefore, perceived management commitment to quality and supervisory commitment

to quality are included to investigate which has a greater impact on individual


Organizational level factors. The final two hypothesized predictors are perceptions of

climate: an improvement in quality climate and commitment to improvement at the

site. The concept of climate has its roots in Lewin's (1951) field theory and later formed

the foundation of the early work of Human Relations. It has been defined as sets of

perceptually based descriptions of relevant organizational features, events and

processes (James and Jones, 1974; Jones and James, 1979). Perceptions of climate

mediate the relationship between organizational context and individual responses,

127

providing a basis for behaviour and affect (Schneider, 1983a, 1983b). Early theorists

(Blake and Mouton, 1964; Likert, 1967) regarded leadership as an important factor in

affecting climate perceptions. Contemporary research (Kozlowski and Doherty, 1989)

has found that leader-subordinate interaction mediates and structures subordinate

interpretations of climate. The authors due to the cross sectional nature of their data did

not examine the possibility of reciprocal linkages nor did they link climate to individual

behaviour. However, Scott and Bruce (1994) found that perceived organizational

support for innovation was positively related to individual innovative behaviour. Here,

we hypothesize that the degree to which organizational members perceived an

improvement in quality climate and commitment to improvement would affect their

commitment to quality and improvement.

6.5 Measures

As indicated in the previous chapter, where possible, previously validated measures

were used. This section presents the additional measures used in the analysis that have

not been previously discussed. The alpha coefficients are given for the combined site 1

and site 2 sample.

It is important to note that the questions tapping general orientation to quality and

improvement as part of the job were asked only at time 2. However, in order to assess

change over time, respondents were asked to give their response to a particular item and

then asked to think back a year prior and asked how they think they would have

responded to the same item.

A general orientation to quality This four item scale, as we have seen, taps an

individual's willingness to exert effort to prevent mistakes and search for

128

improvements. This scale has an alpha coefficient of .70 retrospective for time 1 and

.71 at time 2.

Improvement as part of the job As already noted, this three item measure taps the

degree to which individuals feel improvement is part of their job. The alpha coefficient

for this scale was .61 retrospective for time 1 and .60 for time 2.

Intrinsic motivation This is a six item scale developed by Wan, Cook and Wall

(1979). This scale taps the degree to which a person wants to work well in his/her job

in order to achieve intrinsic satisfaction. The alpha coefficients for this measure were

.77 at time 1 and .84 at time 2.

Higher Order Need Strength This five item scale was developed by Wan Cook

and Wall (1979) to tap an individual's need for satisfaction and achievement through

work. The construct is conceptually distinct from intrinsic motivation which refers to a

specific job situation while this scale is viewed as a dispositional characteristic

extending across jobs. This scale was found to be factorially independent from the

scale measuring intrinsic motivation. The alpha coefficient for this measure was .83 at

time 1 and .84 at time 2.

Organizational commitment This six item scale was adapted from Cook and

Wall's (1980) nine item measure. The three items that were omitted were negatively

phrased. Previous analysis of the scale has shown the six item version to be

psychometrically superior (Peccei and Guest, 1993). The measure contains two items

that tapped each of the three components of organizational commitment; (i)

identification: pride in the organization and internalization of the goals of the

organization; (ii) involvement; willingness to exert effort on behalf of the organization

129

and; (iii) loyalty; attachment to the organization and a desire to remain in the

organization. The scale exhibited high levels of reliability at time 1 and time 2, .83 and

.84 respectively.

Supervisor commitment to quality This six item measure taps respondents'

perceptions of their immediate supervisor's commitment to quality. Two items tap the

extent to which an individual perceives his/her immediate boss to be committed to

improving quality and setting an example of quality performance in their work. The

remaining four items tap an individual's perception of the quality orientated behaviour

of the immediate boss; the degree to which the immediate boss facilitates quality work

and encourages improvements. The alpha coefficients for this scale was .89 and .87 for

time 1 and time 2.

Improvement in quality climate This eight item scale taps respondents' assessment

of whether there has been an improvement in quality climate at the site. This measure

has three components. The first component taps the degree to which top management

is perceived to be more committed to Total Quality and more supportive of suggestions

for improvement. The second component taps respondents' perceptions of Total

Quality having a greater priority, of progress being made in improvement and of the

extent to which individuals are encouraged to voice improvement ideas at the site. The

last component taps management employee relations; that is, the degree to which

individuals perceive greater contact, communication and co-operation between

management and employees. This variable was found to be factorially distinct from

improvement in commitment to quality at the site. The alpha coefficient for this scale

at time 2 was .93.

130


As noted above, in attempting to measure change in commitment to improvement,

respondents were asked to give their view on an item of commitment to improvement

and directly following, they were asked how they would have responded to the same

statement a year ago. Arguably, this retrospective method of assessing change is

subject to a host of biases and the operation of social desirability; not wanting to be

seen as less committed to improvement in the past. Also, this method of assessing

change is not as methodologically rigorous as independent measurements at two points

in time. This raises the question of whether this type of data (an independent

measurement at one point and a retrospective measurement to a previous point in time)

can be used to represent change that has occurred over time. The first step in the

analysis is to assess the appropriateness of using these data to measure change over

time.

6.6.1 Measurement of change using retrospective data

An initial starting point would be to look at the means and standard deviations of the

two components of commitment to improvement and compare them to other similar

conceptual measures gathered independently at time 1 and time 2. Table 6.3 shows the

means and standard deviations for general orientation to quality, improvement as part

of the job, intrinsic motivation, higher order need strength and organizational

commitment using the combined sample from site 1 and site 2.

131

Table 6.3: Means and standard deviations of general orientation to quality, improvement as part of the job and conceptually similar measures

(N=444)

Time 1

Mean (S.D)

Time 2

Mean (S.D)

General orientation to quality 5.70 (0.77) 5.76 (0.76) Improvement as part of the job 4.94 (1.25) 4.97 (1.26) Intrinsic motivation 6.16 (0.64) 6.15 (0.72) Higher order need strength 5.92 (0.77) 5.93 (0.82) Organizational commitment 5.24 (1.03) 5.24 (1.09)

From this, we can see that general orientation to quality and improvement as part of the

job do not elicit as much social desirability as intrinsic motivation and higher order

need strength. Furtheimore, the standard deviations for general orientation to quality

and improvement as part of the job are at least comparable at both time periods to that

of intrinsic motivation and higher order need strength. For improvement as part of the

job, the standard deviations are comparable to that of organizational commitment.

Consequently, these results support the retrospective measure of commitment to

improvement as being as sensitive to comparable measures in picking up variations

between individuals.

The next criteria would be to look at the sensitivity of commitment to improvement at

picking up changes between time 2 and retrospective to time 1. Again, it is necessary

to compare the results to independent measures at time 1 and time 2 of similar

concepts.

For the overall sample, 39.6% of respondents felt less intrinsically motivated at time 2,

20.3% of respondents felt no change in their intrinsic motivation and 40.1% of

respondents felt more intrinsically motivated. With regard to changes in higher order

need strength, 37.4% of respondents reported less need for satisfaction and achievement

132

through work, 20.7% of respondents felt no change and 41.9% reported a greater need

for satisfaction and achievement through work. 13.1% of respondents reported less

general orientation to quality, 61.7% reported no change while 25.2% reported

increased general orientation to quality. Regarding improvement as part of the job,

11.5% of respondents felt that improvements were part of their job to a lesser extent,

71.6% reported no change while 16.9% reported a more positive view of improvements

as part of the job. These results suggest that the retrospective measures of general

orientation to quality and improvement as part of the job may not be as sensitive as

comparable independent measures at time 1 and time 2 in picking up changes over

time.

The intercorrelations between time 1 and time 2 confirm the problematic nature of

using retrospective questions to assess change. The intercorrelations over time for

intrinsic motivation, higher order need strength and organizational commitment are .59,

.47 and .60 respectively. Over time, general orientation to quality has a very high

intercorrelation of .79 while the intercorrelation for improvement as part of the job is

even higher at .92. Based on the high intercorrelations of general orientation to quality

and improvement as part of the job in comparison to the intercorrelations of other

measures, the results of the analysis using the change data needs to be interpreted with

caution. This does not apply to changes in intrinsic motivation. Consequently, in the

discussion, greater weight will be placed on the cross sectional time 2 results.

The general model of commitment to improvement outlined previously was used to

examine the antecedents of general orientation to quality, improvement as part of the

job and intrinsic motivation using Ordinary Least Squares (OLS) regression. The three

dependent variables were regressed separately on the antecedents and the usual set of

control variables using the cross sectional time 2 data and the change data. The results

133

will show the significant antecedents of the three concepts in addition to assessing the

reasonableness of the model for its use in the subsequent chapter.

6.7 Results

The results are presented for site 1 followed by the results for site 2. For ease of clarity

only the standardized beta coefficients for the significant predictors from the model are

shown. In light of the previous discussion on the measurement of change, greater

emphasis is attached to the cross sectional results.

6.7.1: Site 1 results

Table 6.4 presents the significant antecedents of the three dependent variables: general

orientation to quality, improvement as part of the job and intrinsic motivation.

Table 6.4: The impact of the antecedent variables on the three dependent variables at time 2- site 1

General orientation

Improvement as part of the Intrinsic

Antecedent variables to quality job motivation

Management commitment to quality Supervisor commitment to quality Organizational commitment .17*** .18*** .23*** Higher order need strength .16*** .26*** Quality awareness .34*** .22*** .31*** Improvement in quality climate .13+ Improvement in commitment to quality at the site

Adjusted R2 .49 .53 .38 N 216 216 216

+ = p<.10 ** = p <.05 *** = p < .01

134

As can be seen, four of the model variables had a significant impact in the hypothesized

direction on general orientation to quality. Two and three variables respectively were

found to have a significant impact on improvement as part of the job and intrinsic

motivation. None of the leader level variables were found to be significant in

explaining variations in the three dependent variables. Overall, the individual level

antecedents had a stronger effect on all three dependent variables.

The three dependent variables share two antecedents: organizational commitment and

quality awareness. The results are broadly consistent with the results of the factor

analysis. There is adequate support for separating the two dimensions of commitment

to improvement. General orientation to quality and intrinsic motivation share three

antecedents while general orientation to quality has an independent antecedent. Table

6.5 presents the results of the change data.

Table 6.5: The impact of changes in the antecedent variables on changes in the three dependent variables - site 1

Changes in general

orientation

Changes in improvement as part of the

Changes in Intrinsic

motivation Antecedent variables to quality job

A in mgt. commitment to quality A in sup. commitment to quality .17** .14** .12+ A in organizational commitment .18** A in higher order need strength .17** A in quality awareness Improvement in quality climate .18+ Improvement in commitment to quality at the site .29***

Adjusted R2 .06 .06 .05 N 216 216 216

+ = p<.10 ** = p <.05

*** = p < .01

135

Several points are worth noting from the results shown in Table 6.5. First, the leader

level variable relating to changes in supervisor commitment to quality had a significant

impact on changes in general orientation to quality, improvement as part of the job and

intrinsic motivation. In addition, the climate variables were found to be significant in

predicting changes in the two dimensions of commitment to improvement. In terms of

the remaining antecedents of changes in intrinsic motivation, these were individual

level variables.

The results of the change data provide stronger support for separating the concepts of

commitment to improvement and intrinsic motivation. However, this may be largely a

result of how the concepts were measured, specifically, the use of a retrospective time 1

measure for the two dimensions of commitment to improvement.

Are the results found here supported using a different sample and a different

organizational context? The next section presents the results of the same analysis using

the site 2 data.

136

6.7.2: Site 2 results

Table 6.6 shows the significant predictors from the intervening and control variables for

the three dependent variables at site 2.

Table 6.6: The impact of the antecedent variables on the three dependent variables at time 2 - site 2

General orientation

Improvement as part of the Intrinsic

Antecedent variables to quality job motivation

Management commitment to quality -.14+ Supervisor commitment to quality .11+ Organizational commitment .26*** .24*** .27*** Higher order need strength .25... .13** .23... Quality awareness .29*** .16** .26*** Improvement in quality climate Improvement in commitment to quality at the site .15*.

Adjusted R2 .47 .41 .32 N 228 228 228

+ = p<.10 ** = p <.05 *** = p < .01

The results show that four of the model variables had a significant impact in the

hypothesized direction on general orientation to quality. The four significant

antecedents came from the individual and organizational level categories. This

replicates the findings of site 1. There was also consistency across the two sites in the

significance of the individual level variables in predicting improvement as part of the

job. However, a divergence appears between the two sites in the antecedents of

intrinsic motivation. In addition to the significant individual level variables found at

site 1, the leader level variables were found to be significant at site 2. In specific terms,

137

supervisor commitment to quality had a positive impact while management

commitment to quality had a negative effect.

In terms of assessing the similarity of the three concepts, there is some support for

treating general orientation to quality and improvement as part of the job as separate

concepts. There appears to be greater support than found at site 1 for treating intrinsic

motivation as distinct from the two dimensions of commitment to improvement. Table

6.7 presents the results of the same analysis using the change data.

Table 6.7: The impact of changes in the antecedent variables on changes in the dependent variables - site 2

Changes in general

orientation

Changes in improvement as part of the

Changes in intrinsic

motivation Antecedent variables to quality job

A in mgt. commitment to quality A in sup. commitment to quality .20*** .17** A in organizational commitment .20*** A in higher order need strength -.11+ _ . 17*** .19*** A in quality awareness .23*** Improvement in quality climate .20** .31*** Improvement in commitment to quality at the site .18**

Adjusted R2 .17 .14 .09 N 228 228 228

+ = p<.10 ** =p <.05 *** = p < .01

The two dimensions of commitment to improvement have significant antecedents from

the individual, leader and organizational level categories. In comparison, intrinsic

motivation is unaffected by the leader category. In terms of comparison to the results

of site 1, the leader and organizational categories are common in predicting changes in

138

the two dimensions of commitment to improvement. The individual level variables are

common across the two sites in predicting changes in intrinsic motivation. In place of

the leader category at site 1, the organizational level category is significant at site 2.

However, the impact of changes in higher order need strength on changes in the two

dimensions of commitment to improvement casts doubt on the appropriateness of

measuring change using retrospective questions. The impact is in a negative direction

as opposed to the positive impact of higher order need strength on changes (measured

independently at time 1 and time 2) in intrinsic motivation. Together these results

question the validity of using retrospective data. However, a similar result was not

found at site 1. Furthermore, the impact of changes in supervisor commitment to

quality and organizational commitment have a positive impact on changes in

improvement as part of the job.

Support was found for treating the two dimensions of commitment to improvement and

intrinsic motivation as separate. The subsequent discussion focuses on three main

issues. First, the antecedents of the three dependent variables are discussed. This is

followed by an examination of the empirical support for treating the three concepts as

distinct. Finally, the appropriateness of the model for evaluating the impact of the TQM

intervention on the three dependent variables in the subsequent chapter is examined.

6.8 Discussion

In the hypothesized model, the antecedents were categorized in terms of individual,

leader and organizational level variables. The results showed that in explaining

variations in intrinsic motivation, all three of the individual level variables

(organizational commitment, higher order need strength and quality awareness ) were

139

significant across the two sites. Two of these variables: organizational commitment

and higher order need strength at site 1 and higher order need strength and quality

awareness at site 2 are significant in predicting changes in intrinsic motivation. The

impact of higher order need strength on intrinsic motivation is not surprising.

Individuals who have a natural tendency to want satisfaction and achievement through

work are more likely to want to do well in their present job. The impact of

organizational commitment on intrinsic motivation is consistent with previous research

findings (Mowday et al., 1979; DeCotiis and Summers, 1987). Quality awareness also

had a significant impact on changes in intrinsic motivation but only at site 2. None of

the remaining variables in the model had a strong and/or consistent impact on intrinsic

motivation, suggesting that this variable is primarily affected by individual rather than

leader or organizational level factors. 1

Turning to general orientation to quality, there is greater consistency across the two

sites in terms of the significant antecedents. In explaining variations in general

orientation to quality, the individual and organizational level variables are important.

Organizational climate in terms of perceived commitment and support for quality and

improvements was found to be important in affecting an individual's general

orientation to quality. This is consistent with Lewin's (1951) argument that climate is

an important determinant of individual motivation and behaviour. In more recent work,

Scott and Bruce (1994) found that a dimension of perceived climate, support for

innovation, was positively related to individual innovative behaviour. Thus, climate

represents signals an individual receives concerning organizational expectations of

1 Perceived management commitment to quality has significantly shifted in a negative direction between time 1 and time 2. The negative impact of this on intrinsic motivation may be due to individuals externalizing the consequences of the changes in the sense that management are viewed as less committed to quality while individuals intrinsic motivation has remained unchanged.

140

behaviour and gives cues as to what is viewed as important by the organization.

Consequently, individuals adapt their behaviour to reflect organizational expectations.

While the rationale provided is applicable to general orientation to quality, it raises the

question of why a similar effect is not visible for intrinsic motivation. There has been a

suggestion that many types of climate exist (Schneider, 1975) but to speak of climate

without attaching a referent is meaningless (Schneider and Reichers, 1983). In this

study, the climate of interest was support and commitment for quality. Consequently, it

may not have very much influence on an individual's desire to work well to achieve

intrinsic satisfaction. Also, it may be more difficult for perceptions to affect individual

motivation if individuals are already highly motivated. In other words, perceptions of

climate may have an impact on shaping particular behaviours or attitudes in the

formative stage but once a particular level is reached, the climate loses its impact.

In terms of the impact of leadership on general orientation to quality, the evidence is

consistent between the two sites. First, while there is great emphasis placed on

management commitment to quality in the literature, this variable was not found to

have a significant direct effect on an individual's attitude to quality. This is not to say

that it is unimportant as supervisors' behaviour and attitudes may be key indicators of

managerial attitudes and behaviour and supervisors look to the level above for cues as

to what is important. But, in terms of psychological and physical proximity, the

immediate supervisor is closer to the individual and consequently will have a more

salient effect.

The importance of supervisory commitment to quality in affecting improvement as part

of the job is highlighted by the two sites. This is consistent with the Pygmalion effect

(Livingston, 1969; Eden, 1984) in that supervisory behaviour communicates to

141

subordinates the expectations of the supervisor. Scott and Bruce (1994) found support

for this in the context of innovation. Also, the results support Lewin's (1943) concept

of psychological proximity in that the immediate supervisor is more proximal to the

individual than management and consequently, the supervisor should have a greater

influence on individual attitudes and behaviour. These results highlight the importance

of the cascading approach of a TQM change intervention. In order to affect employees'

attitude to quality and improvement, it is not sufficient for management to be

committed to quality as the chain of influence is not a direct one. Consequently, the

supervisor has a pivotal role to play in linking management to employees. This is

consistent with Klein's (1984) argument on the importance of supervisory support to

affect change at the employee level.

To summarise, in broad terms, the significant antecedents of intrinsic motivation are

individual level variables. In contrast, the two dimensions of commitment to

improvement are influenced not only by individual level variables but also leader and

organizational level variables. Thus, the model, in terms of its explanatory power, is

better at explaining general orientation to quality and improvement as part of the job

than intrinsic motivation.

Overall, there appears to be a subtle difference in the effect of factors external to the

individual in affecting the two dimensions of commitment to improvement and intrinsic

motivation. Table 6.3 showed that intrinsic motivation elicited a higher score than

general orientation to quality and improvement as part of the job (6.16, 5.70 and 4.94 at

time 1 respectively). One possible explanation, although purely speculative, is that,

once a particular level is reached, external factors lose their impact. For example,

intrinsic motivation is very high at both sites (6.22 at site 1 and 6.09 at site 2) so

consequently, intrinsic motivation may be impervious to a large degree from the effects

142

of external forces. In comparison, general orientation to quality and improvement as

part of the job may be more easily affected by factors external to the individual as these

attitudes may be under the threshold and consequently more susceptible to the

influences of external factors.

Turning now to examining the degree of similarity between the three concepts in terms

of their significant antecedents, Table 6.8 shows the significant antecedents of the three

concepts for the two sites at time 2.

Table 6.8: Significant antecedents of general orientation to quality, improvement as part of the job and intrinsic motivation at site 1 and site 2 at time 2.

Time 2 results General orientation to quality

Improvement as part of the job

Intrinsic motivation

Antecedents Site 1 Site 2 Site 1 Site 2 Site 1 Site 2

Management commitment to quality ✓ Supervisor commitment to quality ✓ Organizational commitment ✓ ✓ ✓ ✓ ✓ ✓ Higher order need strength ✓ ✓ ✓ ✓ ✓ Quality awareness ✓ ✓ ✓ ✓ ✓ ✓ Improvement in quality climate ✓ Improvement in commitment to quality ✓

The most noticeable result is that in broad terms, the three dependent variables share

the individual level antecedents: quality awareness, higher order need strength and

organizational commitment. General orientation to quality is distinctly affected by

organizational level variables. While these results present a more consistent picture of

the antecedents of the three dependent variables, they do not go so far as to negate the

proposition that these are indeed different concepts. Prior to pursuing this argument,

Table 6.9 presents the significant antecedents of changes in the three dependent

variables.

143

Table 6.9: Significant antecedents of changes in general orientation to quality, improvement as part of the job and intrinsic motivation at site 1 and site 2

Change over time General orientation to quality




Management commitment to quality Supervisor commitment to quality ✓ ✓ ✓ ✓ ✓ Organizational commitment ✓ ✓ Higher order need strength ✓ ✓ ✓ ✓ Quality awareness ✓ Improvement in quality climate ✓ ✓ ✓ Improvement in commitment to quality ✓ ✓

Table 6.9 provides greater support for treating the three variables as distinct as there are

fewer common antecedents between the three variables. However, the significance

attached to these results must be reduced in view of the different methods used to

measure change between the two dimensions of commitment to improvement and

intrinsic motivation.

Overall, the factor analysis provides stronger evidence for treating the three concepts as

distinct as compared to the results of the regressions. Factor analysis is widely used as a

method of investigating the independence of constructs being measured. Thus, while

two concepts may be independent, it is quite possible that they may be influenced by a

substantial number of common predictors.

Putting aside the assessment of independence of concepts using factor analysis, in terms

of predictors, what criteria should be used in assessing whether two constructs are

independent? At one extreme, if one finds that two concepts have the same antecedents

from a given set, what could be inferred from this regarding their independence? On

one hand, one could argue that this would question the independence of the two

concepts. However, one could also argue that the two concepts may have separate

144

antecedents that are not included in the particular group of antecedents chosen. In this

case, one could argue that the three concepts may be measuring an underlying desire to

perform well at work and that they may be measuring distinct aspects or facets of this.

Consequently, one may expect the concepts to share a number of antecedents or to be

influenced by a number of common factors.

Rather than relying totally on subjective judgment to interpret the degree of

independence between the concepts based on their predictors, a crude guideline may

help. This involves taking two independent constructs such as intrinsic motivation and

higher order need strength (Cook et al., 1981) and then looking at the number of shared

antecedents. This will provide a guideline, however crude, for interpreting the results

presented in Table 6.8. The rationale for choosing intrinsic motivation and higher order

need strength is that these two constructs have been measured here and previously

shown to be independent (Cook et al., 1981). Intrinsic motivation and higher order

need strength have a correlation of .40 while the average correlation between the three

concepts (general orientation to quality, improvement as part of the job and intrinsic

motivation) is .44. Consequently, as the correlations are comparable, the number of

shared antecedents between intrinsic motivation and higher order need strength

provides a reasonable guideline. This would not be the case if the correlation between

intrinsic motivation and higher order need strength was radically different from that of

the three concepts. Intrinsic motivation and higher order need strength are regressed

separately on a set of antecedents2 in order to determine how many antecedents they

share.

2 The variables included as predictors of intrinsic motivation and higher order need strength were: management commitment to quality, supervisor commitment to quality, quality awareness, organizational commitment, intrinsic job satisfaction, an improvement in quality climate and an improvement in commitment to quality at the site. Also, a series of control variables were included.

145

The results (Appendix 14) from site 1 showed that intrinsic motivation and higher order

need strength shared two antecedents with the latter construct having one separate

antecedent. The results from site 2 showed that the two constructs share two

antecedents with intrinsic motivation having an additional antecedent. Overall, the

results support the contention that while two concepts may be factorially independent,

they may be influenced or affected by a substantial number of common antecedents.

Using this as a rough guideline, there is support for treating general orientation to

quality, improvement as part of the job and intrinsic motivation as independent; that is,

the concepts are measuring distinct facets of an individual's desire to work well.

The next issue that needs to be addressed is the consistency of significant antecedents

between the two sites. This is based on the assumption that the greater the consistency

of predictors in two different organizations and organizational contexts, the more robust

the model. Referring to Table 6.8, there is a great degree of consistency in predicting

general orientation to quality and improvement as part of the job. In comparison, while

there is consistency in the antecedents from the individual level category in predicting

intrinsic motivation at the two sites, site 2 has an independent influence from the leader

category. Similarly, from Table 6.9, there is greater consistency between the two sites

in predicting changes in the two dimensions of commitment to improvement than

changes in intrinsic motivation. Overall, in view of the different organizational

contexts, there is a reasonable degree of consistency between the two sites with greater

consistency in predicting variations rather than changes in the three dependent

variables.

The use of intrinsic motivation and commitment to improvement in the literature may

reflect different dimensions of employee behaviour in the context of TQM. Intrinsic

motivation, general orientation to quality and improvement as part of the job may be

146

conceptualized roughly in teens of a hierarchical model of employee behaviour. The

difference lies in the degree of effort and commitment required from the individual.

For example, it seems straightforward in terms of effort, to have employees call

attention to problems they have in trying to do their job well. This is similar to

Deming's (1986) prescription for management in that they must remove obstacles that

prevent individuals from gaining satisfaction from their work (intrinsic motivation).

However, regarding improvement as part of the job, one could argue that this requires

greater effort from the individual to treat improvements as an integral part of their job.

Finally, the remaining issue relates to how appropriate the model is to examining the

impact of the TQM intervention on commitment to improvement in the next chapter.

This involves an assessment of how reasonable the model is in predicting the three

variables. Table 6.10 combines the significant antecedents of variations in the three

variables (Table 6.8) and the significant antecedents of changes in the three variables

(Table 6.9)

Table 6.10: Summary of the significant antecedents of variations and changes in general orientation to quality, improvement as part of the job and intrinsic motivation at site 1 and site 2.

General orientation to quality




Management commitment to quality X Supervisor commitment to quality X X X X X X Organizational commitment X X X X X X Higher order need strength X X X X X Quality awareness X X X X X X Improvement in quality climate X X X X Improvement in commitment to quality X X X

147

Overall, there is empirical support for the model and between the three dependent

variables, all the model variables were found to be significant. However, one could

question the validity of retaining management commitment to quality as a model

variable given its limited effect. The rationale for its inclusion outlined earlier was to

investigate the impact of management commitment to quality in relation to supervisor

commitment to quality. In order to be consistent between the testing of the antecedent

model and using it as a basis for evaluation, management commitment to quality will

be retained as a model variable in the subsequent chapter.

6.9 Conclusions

This chapter set out to investigate whether commitment to improvement and intrinsic

motivation were the same concepts and thus interchangeable as evidenced in the

literature. The results of the factor analysis together with the results of the regressions

provide reasonably strong evidence that commitment to improvement is not

unidimensional. Furthermore, support was found for treating the two dimensions of

commitment to improvement and intrinsic motivation as independent constructs.

In terms of the overall model, there was a reasonable degree of consistency between the

two sites in predicting variations in the three dependent variables. In addition, the

outlined model contains antecedents that are reasonably good at predicting the three

dependent variables. Having assessed the independence of the three concepts and the

reasonableness of the model, the next chapter investigates the impact of the TQM

intervention on each of the three concepts.

148

Chapter 7: The Impact of a TQM Intervention on Commitment to Improvement and Intrinsic Motivation

7.1 Introduction

The previous chapter examined the nature of commitment to improvement and intrinsic

motivation. The empirical evidence suggested that commitment to improvement was

not a unidimensional construct. Rather, it consists of two dimensions: general

orientation to quality and improvement as part of the job. In addition, there was

empirical support for treating the two dimensions of commitment to improvement and

intrinsic motivation as independent constructs. Based on these findings, this chapter

sets out to investigate the impact, if any, of the TQM intervention at site 1 on each of

the three concepts separately: general orientation to quality, improvement as part of the

job and intrinsic motivation.

This chapter may be thought of as an extension of chapter 5 which assessed the impact

of the TQM intervention on team orientation. The results indicated that participation in

the intervention did not have a significant effect on team orientation. One aspect of the

intervention, supervisory reinforcement was found to have a significant total effect on

team orientation. To what extent are the findings of chapter 5 replicated with respect to

the other key outcomes of TQM; general orientation to quality, improvement as part of

the job and intrinsic motivation?

This chapter begins by looking at the evaluation model followed by the descriptive

statistics which examine the extent and direction of change in the model variables.

Subsequently, the results are presented and discussed.

149

7.2 Evaluation model

The model of the antecedents of the two dimensions of commitment to improvement

and intrinsic motivation outlined in the previous chapter provides a basis for assessing

the impact of the TQM intervention on general orientation to quality, improvement as

part of the job and intrinsic motivation. A diagram outlining the full evaluation model

(containing the intervention variables) is depicted in Figure 7.1. Once again, the

intervention is assessed in teems of its total impact on the three dependent variables.

This involves assessing the direct and indirect effects of the four intervention variables

on each of the dependent variables.

The previous chapter concluded that the 'antecedents' model was reasonably good in

predicting the three dependent variables. A rationale was presented for retaining

management commitment to quality in the model given its limited effect. However, one

could question why, for example, higher order need strength is included as a predictor

of improvement as part of the job and similarly, why improvement in quality climate is

included in the intrinsic motivation model, as they were not found to be significant

predictors when the model was tested in the previous chapter. The rationale for using

the same predictors for the three dependent variables is to allow a comparison of the

impact of the TQM intervention across the three variables.

150

INS

E a)

1=4

a)

E

151

Com

mitm

ent t

o im

prov

emen

t

The evaluation model shown in Figure 7.1 was tested using OLS regression in two

stages. The first stage involved regressing the intervening variables (the antecedents)

on the control and the intervention variables. In the second stage, the three dependent

variables were then regressed separately on all the variables in the model. Similar to

the previous chapter, the analysis was conducted on the time 2 cross sectional data and

on the change data. Prior to presenting the results, the descriptive statistics look at the

extent and direction of change in the model variables at site 1.

73 Descriptive statistics

Table 7.1 presents the results of the paired sample t-tests for the group of employees at

site 1.

152

Table 7.1: Paired sample t-tests for employees at site 1

Variables

Sample (N=165) Employees

Time 1 Time 2 Change Scores

Mean (S.D) Mean (S.D) Mean (S.D)

General orientation to quality 5.62 (0.83) 5.78 (0.75) .16 (0.46)*** Improvement as part of the job 4.45 (1.18) 4.58 (1.19) .13 (0.53)*** Intrinsic motivation 6.20 (0.65) 6.24 (0.68) .04 (0.60) Higher Order Need Strength 5.87 (0.81) 5.88 (0.92) .01 (0.87) Quality awareness 5.99 (0.66) 5.95 (0.65) -.04 (0.59) Organizational commitment 5.32 (0.98) 5.51 (0.99) .19 (0.72)*** Management commitment to quality

5.10 (1.01) 5.27 (1.03) .17 (0.86)**

Supervisory commitment to quality

4.91 (1.25) 5.04 (1.19) .13 (1.12)

Improvement in commitment to quality

4.84 (1.06)

Improvement in quality climate 4.63 (1.12) Participation in the intervention 2.58 (1.15) Perceived benefit of intervention 4.20 (1.31) Appropriateness of intervention 4.99 (1.26) Reinforcement of intervention 4.55 (1.27)

** T-test difference between Time 1 and Time 2 significant at < than .05 level *** T-test difference between Time 1 and Time 2 significant at < than .01 level

Overall, with the exception of quality awareness, there has been a general shift in a

positive direction. However, this has not been significant in all measures. In terms of

the two dimensions of commitment to improvement, there has been a significant

positive shift. In contrast, there has not been a parallel change in intrinsic motivation.

This may be partly due to the difference in the measurement of the concepts; that is, the

independent measurement of intrinsic motivation at time 1 and the retrospective

measure to time 1 of the two dimensions of commitment to improvement. In addition,

organizational commitment and perceived management commitment to quality have

significantly shifted in a positive direction.

153

7.4 Results

The results of the cross sectional time 2 data are presented first followed by the results

of the change data. First, the impact of the TQM intervention on the intervening

variables is presented. This is followed by the impact of the intervention on the three

dependent variables (this is a test of the full model showing the direct, indirect and total

effects of the intervention on the three dependent variables). Similar to the previous

results chapters, the standardized beta coefficients are shown.

Table 7.2 shows the results of the impact of the intervention and the usual set of control

variables (not reported) on the intervening variables.

154

Tabl

e 7.

2: D

irec

t eff

ects

of th

e TQ

M in

terv

entio

n on

inte

rven

ing

vari

able

s in

the

mod

el a

t tim

e 2

Impr

ove

in

com

mit

to

qual

ity

71- O N ,-.. . * . Cl

in cr, 1/4, ci ,__,

Impr

ove

in

qual

ity

clim

ate +

N * 00 •--, ._,

* 00 ,__,

* cn cry

N kr) 71. k.0

--1

Mgt

. co

mm

it to

qu

ality

C) N 7t. N N N ,—,

Sup.

com

mit

to q

ualit

y

00 O

,..0 0

kr) ci ■0

com

mit.

co

mm

it.

00 C

1,-) C

71- cr?

N C'

c) ,r, Cl 1/40 ,--•

Qua

lity

awar

e.

N 0 N C

kr) —,.

0 . co kr) 0 • .

Hig

her o

rder

ne

ed st

reng

th

0 0

In C'. I

+ 0 N

+ N ---■

N kr) 0 • .

TQM

Var

iabl

es

0 0 0 0 0 0 cc 0 4a' o t

•,-J c.)

a.) 0 a) ,j N a) 0 E 0 c+-1 (I-1 cu (4-■ c+-■ LI-4 C-) r, W W cd W ;..■ 1-.-1 <•1

124 Z E-1 8 iL-4 c,-,) sa■ 0

cu 1-• em

+.4 sa, 0 Qa) a:I Sa• 1=-: ._

0 "0 ct ..--• -0 -0 0

Ei ... C.) • -■ it• -• a) U c_.) U ;-• ta•

ct a) a) c2.1 11.1 f:1•1 Ci)

,--I C,i C,-; .71:

O V

* *

kr) C'. V

* Ca4

O

1i

155

Several points are worth noting from Table 7.2. Overall, participation in the

intervention does not have a significant effect on the intervening variables

(participation has an effect on improvement in quality climate, significant at 10%

level). This parallels the finding of chapter 5 on the lack of significant impact of

participation. In contrast, the assessment of the intervention has a significant positive

impact on the majority of the intervening variables.

Tables 7.3 presents the total impact of the TQM intervention on the three dependent

variables.

Table 7.3: Direct, indirect and total effects of the TQM intervention on the dependent variables at time 2

TQM Variables


Improvement as part of the job Intrinsic motivation

1. Participation in TQM Direct Effect on: .08 -.03 .02

Indirect Effect on: .03 .04 .03 Total Effect: .11 .01 .05

2. Perceived benefit Direct Effect on: -.04 .28*** .13

Indirect Effect on: .07 .06 .01 Total Effect: .03 .34*** .14+

3. Perceived appropriateness Direct Effect on: .08 -.11 -.11

Indirect Effect on: .18 .09 .18 Total Effect: .26*** -.02 .07

4. Supervisory reinforcement Direct Effect on: -.14+ -.09 .02

Indirect Effect on: .19 .14 .10 Total Effect: .05 .05 .12

Adjusted R2 .49 .46 .39 N 165 165 165

+ = p<.10

** = p<.05

*** = p<.01

156

Participation in the intervention does not have a significant direct or indirect impact on

the three dependent variables. In other words, participants are no more likely to be

committed to improvement or intrinsically motivated than non participants. Similarly,

supervisory reinforcement of the intervention was not found to have a significant effect

on general orientation to quality, improvement as part of the job and intrinsic

motivation. However, perceived benefit of the intervention has a significant positive

effect on improvement as part of the job and to a lesser extent on intrinsic motivation.

Those individuals who perceive the intervention as providing benefit are more likely to

be intrinsically motivated and to see improvements as an integral part of their job.

Finally, perceived appropriateness has a significant positive effect on general

orientation to quality.

Table 7.4 presents the results of the impact of the TQM intervention and the control

variables (not reported) on changes in the intervening variables.

In terms of the impact of the intervention on changes in the intervening variables,'

supervisory reinforcement appears to have the greatest effect. Supervisory

reinforcement of the intervention was found to have a significant effect on changes in

higher order need strength, organizational commitment and perceived supervisor

commitment to quality. Neither perceived benefit nor perceived appropriateness were

found to have a significant effect on changes in the intervening variables. Participation

in the intervention was found to have a mild positive effect on changes in

organizational commitment (significant at 10% level).

1 An improvement in quality climate and an improvement in commitment to quality were measured at time 2 only and thus they do not represent changes over time.

157

Dire

c t e

ffect

s of t

he T

QM

inte

rven

tion

on in

terv

enin

g va

riab

les

in t

he m

odel

- ch

ange

ove

r tim

e (T

2-T1

)

E-(

Impr

ove

in

com

mit

to

qual

ity

-71. 0 d • (5., —

kc) 01 1/4,0 CI .-.1

Impr

ove

in

qual

ity

clim

ate

+ N r.-*

• * 00 ..-■

* 00 ,-1

* * *

en cry N kr) 71- VD •-•-■

Mgt

. co

mm

it to

qu

ality

kr) ,-, •-, --, en I .

c:r■ ken kr) r.

Sup.

com

mit

to q

ualit

y

en c) N .-■ en ,-:

+ co N cc)

Org

. com

mit.

+ N •--,

71- c)

71- •-, .

1

* ■.o N

kr::: v-, C? 0

,.

Qua

lity

awar

e.

,.--, ,-,

1/40

I O .

kr) o

C) ,n o

Hig

her o

rder

ne

ed s

treng

th

N O

.

co c)

.

N. o

■

-,d)-

(1 v.) ,n c) \ -0 • ,--,

TQM

Var

iabl

es

t o .4., U a)

„....,,,'-' V-) ,..., .4-J H .) C.)

. a)

0 •r. 4..,

•I-. U •r.

cc: A*

o .I.J C.) a)

4-1 W 4-,

1** a) a) . 0 r-N a) :-: ,0 "ti

•a5 c.) )-, a)

fn-: c‘i

;4

v) 0 Cn A-, (I.) c.) 0 a.) c..) r!-■ -1-■ *-* ct W •** ;•-■ 4-.■

0 0 ;-* . f=1. n 0.4 ,--, al

77j

›. •.. a.) c..) N a)

In-: cn

4., o (1) *-' C.) E (1) c..) Lr W CD ±,

.5 a) a) . ).-■ c:2)

... CD

•,-.

a) s=1. 0

ceD 4

c, t: Z ba.) -*-.

"0

V

II

158

Table 7.5 presents the direct, indirect and total effect of the TQM intervention on

changes in the three dependent variables.

Table 7.5: Direct, indirect and total effects of the TQM intervention on the dependent variables- change over time (T2-T1)

TQM Variables


Improvement as part of the job Intrinsic motivation

1. Participation in TQM Direct Effect on: -.02 -.09 -.04

Indirect Effect on: .01 .02 .03 Total Effect: -.01 -.07 -.01

2. Perceived benefit Direct Effect on: -.28*** -.18+ .11

Indirect Effect on: .12 .11 .03 Total Effect: -.16.. -.07 .14**

3. Perceived appropriateness Direct Effect on: .05 .08 -.09

Indirect Effect on: .06 .05 -.04 Total Effect: .11 .13+ -.13+

4. Supervisory reinforcement Direct Effect on: .12 .01 -.09

Indirect Effect on: .14 .15 .13 Total Effect: .26*** .16+ .04

Adjusted R2 .13 .08 -.01 N 165 165 165

+ = p‹.10

** = p<.05 *** = p<.01

Similar to the cross sectional results, participation in the intervention did not have a

significant effect on any of the three dependent variables. Perceived benefit had a

significant positive effect on intrinsic motivation and a parallel negative effect on

general orientation to quality. Perceived appropriateness had a mild positive effect

(significant at 10% level) on improvement as part of the job and a mild negative impact

159

(significant at 10% level) on intrinsic motivation. Supervisory reinforcement was

found to have a significant positive effect on general orientation to quality and to a

lesser extent on improvement as part of the job.

7.5 Discussion

The format of the discussion parallels the results; the impact of the intervention

variables on the intervening variables is the basis of the initial discussion which is

followed by a consideration of the impact of the intervention on the three dependent

variables. As noted earlier, greater weight will be attached to the cross sectional

results in the discussion.

Intervening variables

Overall, the TQM intervention was found to have a significant impact on the

intervening variables. The impact of participation was very limited and this parallels

the finding in chapter 5. In contrast, the two assessment variables and supervisory

reinforcement of the intervention had a greater impact on the intervening variables. We

begin by looking at the impact of the intervention on the individual level variables

before moving on to the impact of the intervention on the leader and organizational

level variables.

One individual level variable, namely, quality awareness remained unaffected by the

intervention. 2 The impact of the intervention on higher order need strength is primarily

due to supervisory reinforcement of the intervention. 3 This effect is potentially due to

the actions and communications of the supervisor. By involving subordinates in the

2 The lack of significant effect of the intervention on quality awareness was discussed in chapter 5. 3 Supervisory reinforcement had a significant effect on changes in higher order need strength (Table 7.4) and also a mild effect on variations in higher order need strength (Table 7.2- significant at 10% level)

160

intervention, the supervisor may be enhancing their need for achievement and

satisfaction through work. In addition, the supervisor may communicate to his/her

subordinates the potential of the intervention to give them greater influence and to

extend their abilities thus raising the importance individuals attach to the fulfillment of

higher order needs.

Individuals who perceive their supervisor as reinforcing the intervention are also more

likely to experience greater commitment to the organization. By reinforcing the

intervention, supervisors are involving subordinates and allowing them to participate in

the improvement process. This is consistent with the links reported in Jermier and

Berkes (1979) and Rhodes and Steers (1981) between participatory leadership and

organizational commitment. In addition, at time 2, the more individuals perceive the

intervention as appropriate, the more committed they are to the organization. However,

it is plausible that this relationship operates in the opposite direction; that is, that the

more individuals are committed to the organization, the more likely they are to view the

intervention as appropriate. As this was found in the results of the cross sectional time

2 data, it is not possible to discern which direction of causation holds true. This is

further investigated in chapter 9 which examines the predictors of perceived benefit and

perceived appropriateness of the intervention.

The impact of supervisory reinforcement on perceived supervisor commitment to

quality is not surprising. This is because reinforcing the intervention is a key

behavioural indicator of a supervisor's commitment to quality. The impact of

supervisory reinforcement of the intervention on perceived management commitment to

quality highlights the importance of the immediate boss. This is in line with Kozlowski

and Doherty's (1989) argument. They argue that processes which characterize the

interactions within the immediate organizational context are expected to have much

161

closer links to perceptions. Consequently, leadership behaviours of immediate

supervisors are likely to be representative of wider organizational processes.

Subordinates who perceive their immediate boss as reinforcing the intervention and

therefore being more committed to quality will tend to generalize their perceptions of

the supervisor to the organization in general (i.e. management). Further support for this

is found in the two organizational climate perceptions where supervisory reinforcement

of the intervention was found to be the strongest predictor.

The intervention has a significant impact on individuals' perception of an improvement

in quality climate and in commitment to quality at the site. The more positive an

individual perceives the intervention to be, the more positive the perceptions of

organizational climate. This is not surprising given that organizational climate

represents "perceptually based descriptions of relevant organizational features, events

and processes" (James and Jones, 1974; Jones and James, 1979). Therefore, an

individual who perceives their supervisor to be reinforcing the intervention and who

perceives the intervention as appropriate and beneficial is more likely to perceive an

improvement in quality climate and commitment to quality at the site. Once again, it is

not an individual's participation in the intervention per se that affects their climate

perceptions but rather the assessment of the intervention

To summarize, the intervention was found to have a significant effect on the

intervening variables. However, the impact varied from having no significant impact

on an individual's quality awareness to having an extensive impact on an individual's

perception of an improvement in quality climate. We now turn to the overall impact of

the intervention on the three dependent variables.

162


Overall, the intervention has a significant impact on intrinsic motivation. Specifically,

one aspect of the intervention; perceived benefit has a significant positive effect on

changes in intrinsic motivation and a mild positive effect on variations in intrinsic

motivation. Wilkinson et al. (1990, 1991) provide case study evidence which suggests

that TQM may offer tangible benefits to employees. The findings of this study go

further to suggest that if individuals perceive the intervention as providing some benefit

to them, they are more likely to experience greater intrinsic motivation.

The explanation provided by some of the quality proponents would be that perceived

benefit would involve eliminating some of the obstacles that prevent employees from

doing the best job they can. This explanation would involve viewing a TQM

intervention as providing a legitimate mechanism that allows and encourages

employees to voice problems or constraints they have that makes it more difficult for

them to do an efficient or quality job. These inhibitors may include material resources,

physical work-environment variables or work processes. Consequently, by eliminating

obstacles individuals confront in doing their job, the individual is given more control

over his/her performance thereby strengthening the link between effort and

performance which should, in turn, result in greater intrinsic motivation. Fisher (1978)

found support for the link between personal control over performance and intrinsic

motivation.

The measure of perceived benefit is a crude one and consequently, it is not possible to

outline specifically what individuals see as beneficial in terms of what the TQM

intervention provides. Benefit to some individuals may be interpreted as making their

job easier (the elimination of obstacles), while for others, it may be perceived as giving

them an opportunity for greater involvement and greater influence in their work area.

163

Wilkinson and Willmott (1995) note that benefits to employees may include the

removal of excessively close supervision, unreliable services from other departments

and the lowering of barriers between management and employees. Regardless of how

benefit is perceived, there is an element of self interest that seems to influence an

individual's intrinsic motivation.


Perceived benefit of the intervention also has a significant effect on improvement as

part of the job (it should be noted that the direct effect of perceived benefit on

improvement as part of the job is significant). Thus, the greater the perceived benefit of

the intervention, the more an individual perceives improvements as an integral part of

their job. The explanation may be similar to that of intrinsic motivation. Employees

may view the intervention as providing the opportunity for them to make improvements

in their work and how they accomplish it and consequently be more inclined to view

improvements as an integral part of their job.

Although purely speculative, the direction of influence between perceived benefit and

improvement as part of the job may go the other way; that is, that individuals who view

improvements as part of their job are more likely to assess the intervention as

beneficial. Individuals may see the intervention as providing a concrete mechanism by

which improvements may be implemented.

164


Perceived benefit of the intervention was not found to have a significant effect on

general orientation to quality Rather, another aspect of the intervention, its perceived

appropriateness, was found to have a significant effect. In other words, the more

appropriate individuals perceive the intervention to be, the more oriented they are to

quality. In chapter 3, a rationale was put forward for including the direct and indirect

effects of the intervention thereby evaluating the intervention on the basis of its total

impact. A case in point is the impact of perceived appropriateness on general

orientation to quality. If one were to rely on direct effects, from Table 7.3, one can see

that perceived appropriateness does not have a significant direct effect on general

orientation to quality. However, by including indirect effects, perceived

appropriateness has a significant total effect.

The differing impact of perceived benefit and appropriateness on the three dependent

variables is an interesting finding. Perceived benefit has a significant impact on

intrinsic motivation and improvement as part of the job; perceived appropriateness has

a significant effect on general orientation to quality. What this appears to indicate is

that for the intervention to affect an individual's attitude toward their present job, it

must satisfy an individual's self interest by providing some benefit to them. As

mentioned before, the measure of benefit is a crude one and it is quite possible that an

individual may not perceive the same benefit as having an impact on both intrinsic

motivation and improvement as part of the job. The differentiating characteristic of

general orientation to quality is that it does not focus on an individual's job or their

attitude toward their job. It is more altruistically oriented in the sense that it taps a

general willingness to prevent mistakes in the work area and satisfaction from making a

contribution to improving things generally.

165

Perceived appropriateness includes one item that taps benefit to the individual. But

overall, the emphasis is on appropriateness of the intervention as a way of bringing

about the needed change at the site, the perceived priority placed on the intervention as

well as the perception of equal benefits to management and employees. This result

raises the question of why perceived appropriateness positively affects general

orientation to quality but not intrinsic motivation and improvement as part of the job

(significantly affected by perceived benefit of the intervention). One potential

explanation is that individuals have a general orientation to quality (this is an

assumption made by most of the quality proponents) and the intervention provides and

appropriate mechanism to allow this orientation to develop. One could speculate that

the direction of influence may go the other way. Individuals who have a general

orientation to quality may be more likely to see the intervention as an appropriate way

to bring about the change needed at the site. Due to the nature of the data, it is not

possible to ascertain, in this case (as with improvement as part of the job) the direction

of influence.

These findings are subject to a number of caveats. First, the evaluation model assumes

that the direction of influence is from the intervention to the three dependent variables.

In other words, the model assumes that general orientation to quality, improvement as

part of the job and intrinsic motivation are endogenous in relation to the other variables

in the model. This was not challenged here. 4 It is plausible that individuals who are

oriented toward quality to begin with will be more likely to view the intervention as

appropriate. Similarly, individuals who are intrinsically motivated and who view

improvements as an integral part of their job may be predisposed to perceiving the

4 In Chapter 4, the endogeneity assumption was tested using cross lagged regressions. In this chapter, it was not possible to do this in light of the method by which change in general orientation to quality and improvement as part of the job was measured.

166

intervention as providing benefit. Second, although the impact of perceived benefit is

consistent with previous research findings and suggestions, the measure used here is

general and crude. Future studies should consider in specific terms what individuals

see as beneficial. Despite the limitations of these findings, there are practical

implications for the implementation of TQM

Overall, the findings suggest that the intervention has differential effects depending on

the attitude and behaviour being affected. Therefore, if one intends to affect "injob"

attitudes and behaviours ( this includes intrinsic motivation and improvement as part of

the job which focus on individual attitudes toward the job), then the results suggest that

for this to occur, an individual must perceive the intervention as providing benefit to

them. However, perceived benefit, may not be important for affecting an individual's

attitude to quality and improvement in the wider work area. In this case, it is more

important for individuals to see the intervention as appropriate. Consequently, different

aspects of the assessment have different effects on a range of individual quality oriented

attitudes and behaviours with the key distinguishing characteristic being whether these

attitudes and behaviour are inward (toward the job) or outward (toward the work group

or work area).

Finally, the lack of significant effect of participation is consistent with the findings of

chapter 5. Thus, participation was not found to have an important influence on team

orientation, general orientation to quality, improvement as part of the job or intrinsic

motivation. This suggests that it is not what individuals do that is important but rather

what they think. As participation and assessment were measured at the same time, it is

not possible to discern the direction of causation between them. It is quite possible that

participation influences assessment which in turn influences future participation. For

example, an individual who participates in the intervention and makes a positive

167

assessment may desire greater participation. On the other hand, an individual may

make an initial assessment which influences their future participation. With this in

mind, the thesis so far suggests that when participation and assessment are considered

as competing predictors, assessment is more significant in affecting subsequent

attitudes and behaviour.

In terms of the impact of assessment of the intervention, from this chapter it was shown

that perceived benefit and appropriateness affected the dependent variables differently.

In chapter 5, the significant predictor was supervisory reinforcements of the

intervention. If these results are broadly correct, there are implications for the

management of a TQM change intervention. As well as attempting to ensure that the

intervention is perceived as beneficial, it may also be necessary to "manage" how the

intervention is perceived along other dimensions of assessment. However, this

recommendation must be viewed in light of the limitations of this study. First, what

may hold true in the short term may not apply to the longer term. Second, it would be

necessary in future studies to tap a wider range of assessment dimensions. Nonetheless,

different dimensions of assessment seem to have a differential impact on quality

attitudes and behaviour.

7.6 Conclusions

This chapter set out to investigate the impact of a TQM intervention on general

orientation to quality, improvement as part of the job and intrinsic motivation. Overall,

the intervention had a significant total effect on all the variables examined here.

Perceived benefit of the intervention had a significant effect on improvement as part of

the job and intrinsic motivation; perceived appropriateness had a significant effect on

5 This is not an assessment variable in terms of how individuals perceive the intervention per se. Rather, it taps an assessment of the process of implementation.

168

general orientation to quality. Overall, the findings suggest that the intervention has a

differential effect depending on the type of attitude and behaviour being affected.

Participation was not found to have a significant effect on any of the dependent

variables. This result is consistent with the findings of chapter 5. The next two

chapters pick up a different line of investigation by looking at the antecedents of

employee participation in and assessment of the intervention.

169

Chapter 8: Predictors of Employee Participation in a TQM Intervention

8.1 Introduction

Up to this point, the focus has been on assessing the impact of the intervention on some

of the core elements of TQM. Chapter 5 examined the impact of the intervention on

team orientation and showed that the intervention did not have a significant direct effect

on team orientation. However, the intervention was found to have a significant total

effect on team orientation. Chapter 7 showed that the intervention had a significant

total effect on general orientation to quality, improvement as part of the job and

intrinsic motivation. In the case of improvement as part of the job, the intervention did

have a significant direct effect. Overall, the intervention was found to have a

significant total effect on all the core variables in this study.

A key finding of this evaluation is that participation per se did not have a significant

impact on any of the core outcome variables examined here (i.e. team orientation,

general orientation to quality, improvement as part of the job and intrinsic motivation).

Rather, it was individuals' assessment of the intervention that played a significant role

in affecting the core elements of TQM. In particular, supervisory reinforcement of the

intervention had a significant (albeit indirect) effect on team orientation while

perceived benefit and appropriateness of the intervention were found to have a

differential impact on the two dimensions of commitment to improvement and intrinsic

motivation.

An assumption of the previous chapters was that the TQM intervention (participation in

and assessment of) was exogenous in relation to the remaining variables. In other

170

words, participation in and assessment of the intervention were treated as antecedents

rather than consequences of the other variables. This and the following chapter test the

validity of this assumption as one could plausibly argue that attitudes prior to the

intervention may affect an individual's participation in and assessment of the

intervention itself.

Consequently, this and the following chapter signal a change in the line of inquiry from

evaluating the impact of the intervention to examining the predictive power of

antecedent attitudes on individual participation in and assessment of the intervention.

More generally, the overall question we now turn to is what predicts individuals'

participation in the intervention and their assessment of it. We start by examining the

predictors of participation. As we have seen, participation does not have a clear effect

on subsequent attitudes. Nevertheless, an analysis of the factors which affect

participation is important for two reasons. First, comparing the predictors of

participation in and assessment of the intervention, may provide an important insight as

to why assessment is more powerful than participation in predicting subsequent

attitudes. Second, participation has and continues to attract a great deal of academic

attention. Questions such as do workers want participation? To what extent do workers

want participation? What types of outcomes are associated with different participation

strategies? and Who participates? are some of the key areas that have been addressed by

researchers (for example, Hespe and Wall, 1976; Ramsay, 1976; Marchington, 1980;

Wall and Lischeron, 1977). This chapter focuses the last question in the context of a

TQM intervention.

Employee participation is a widely researched area. Generally, the primary focus of

empirical research and a point of considerable debate concerns the outcomes resulting

171

from various participation strategies. ' Far less attention has been given to the

antecedents of participation. In briefly reviewing relevant literatures in this area,

emphasis is placed on research investigating the factors influencing participation in

Quality Circles (QCs) as it is highly related to the form of employee participation in

TQM. In fact, it has been argued that QCs are an inherent component if TQM

(Ishikawa, 1985). However, QCs and TQM differ with respect to the role of

supervisors and the degree of choice which employees may have in participating.

These may have consequences for who participates in TQM and are discussed later.

Rafaeli (1985) suggests that different employees may react more positively to QC

activities than others. The author offers the following factors that may affect an

individual's reaction to QC activities; preferred leadership style, growth need strength

and job involvement. Bruning and Liverpool (1993) suggest that due to the voluntary

nature of QCs, QC members would have a greater desire to participate than

nonmembers. In addition, the authors suggest that there is a possibility that participants

may have different personal characteristics to non participants. They found that QC

participants reported higher levels of desired and perceived actual participation, they

were more highly educated and exhibited lower needs for dominance than non

members. The authors argue that QCs attract different types of individuals. However,

their research methodology was cross sectional in nature and the QCs were in operation

from 3-5 years. Hill (1991b) argues that cross sectional research has shortcomings for

QC research or any employee participation program that is voluntary based.

1 See Coch and French, 1948; Guzzo et al., 1985; Lawler, 1982 and; Miller and Monge, 1986 for evidence in support for participation. Contrary evidence is presented by Locke et al., 1980; Wagner and Gooding, 1987a, 1987b and; Wagner, 1994.

172

In attempting to investigate differences between participants and non participants using

a cross sectional research design, any significant differences may be due not to

membership of QCs but rather due to prior differences that led volunteers to select

themselves for membership. There is evidence that where employee participation

programs are voluntary, based on pre program preferences, self selection of individual

into these programs is not uncommon (Cutcher-Gershenfeld, 1986; Verma and

McKersie, 1987). Graham and Verma (1991) argue that where a self selection process

is at work, it is important to identify the influence of self selection characteristics on

participation-outcome relationship. Griffin (1988) echos a similar message and calls

for additional research to gain insight into who chooses to participate in QCs.

There has been some research within the general arena of participation on the link

between individual characteristics and attitudes toward participation such as age

(Sheppard and Herrick, 1972), bureaucratic orientation (Gordon, 1970), gender and

length of service (Hespe and Wall, 1976 for a review). However, there is a need to

investigate a wider range of characteristics that predispose some individuals to

participate and others not to participate. In participation programs where individuals

exercise choice as to whether to participate or not, characteristics which distinguish

participants from non participants may have an impact on the participation-outcome

relationship.

As Hill (1991b) states " outside the framework of TQM, circles continually run up

against the problem that organizations are not structured to respond to bottom-up

initiatives " (p556). TQM overcomes the problems of QCs by integrating quality

improvement into the existing organizational hierarchy. It does this by adopting a top

down approach to change thus overcoming the problems faced by middle managers to

bottom up initiatives. Middle managers are fully integrated into the quality

173

improvement process thus potentially overcoming the problem faced by QCs, that of

resistance from middle managers (Hill, 1991b; Collard and Dale, 1989). Hill (1991b)

presents evidence that middle managers gained from TQM through greater influence

and involvement from their superiors. TQM involves a cascading process of increased

involvement of various layers of management culminating in the first line

manager/supervisor involving his/her subordinates in the improvement process. Thus,

the involvement of one level in the organization is dependent upon the actions of the

next level. In practice, Hill (1991b) found that a number of managers failed to develop

participative teamworking among subordinates and to pursue improvement efforts.

The emphasis on the cascading approach to involvement in TQM departs from the more

independent free standing mechanism of QCs. This may have implications for the

factors affecting participation in TQM. More specifically, the antecedents of employee

participation may not lie exclusively at the level of individual attitudes and

characteristics (as seems to be the case with QCs) but may also include the behaviour of

individuals in level(s) above.

With this in mind, this chapter explores the predictors of employee participation in a

TQM intervention. Using the empirical work on QCs and the framework of TQM, an

hypothesized model of the predictors of employee participation is outlined and tested.

In the subsequent chapter, this is extended to an investigation into the predictors of

employee assessment of the TQM intervention. This is important in view of the

significant impact of the assessment of the intervention shown in previous chapters.

The following section outlines an exploratory model of the predictors of employee

participation in TQM. This is followed by a discussion of the analysis procedures

adopted. Finally, the results are presented and discussed.

174

8.2 An hypothesized model of employee participation in a TQM intervention

Figure 8.1 presents a diagram of the hypothesized model of employee participation in a

TQM intervention. Four main variables are hypothesized to affect an individual's

participation in the intervention; supervisor participative style, organizational

commitment, higher order need strength and influence gap. A second model is tested

which includes perceived benefit of the intervention in addition to the outlined

predictors. The rationale for testing this second model is presented in the next section

175

FIG

UR

E 8

. 1:

N

H

176

Individual and participation This category includes an individual's attitude

toward the organization, toward work in general and perception of his/her desired and

actual influence in their work area. Organizational commitment is hypothesized to

positively affect an individual's participation in TQM. There are two strands of

thinking in relation to organizational commitment. The first strand views

organizational commitment as an outcome of some form of participation (Griffin, 1988;

Wall et al., 1986; Bruning and Liverpool, 1993). Mathieu and Zajac (1990) argue that

organizational commitment is a useful criterion for a range of organizational

interventions (to include participation) aimed at improving employee attitudes and

behaviour.

The second strand interprets organizational commitment as an antecedent of a range of

behavioural outcomes; extra role behaviour (Katz and Kahn, 1978; Brief and

Motowidlo, 1986), increased effort (Randall, 1990) and Organizational Citizenship

Behaviour (Eisenberger et al., 1990). This view hypothesizes that individuals who are

committed to the organization are more likely, in a variety of ways, to exert effort on

behalf of the organization. Following from this, given the opportunity to participate in

work related matters, individuals with greater commitment to the organization are more

likely to take the opportunity to participate.

The first perspective assumes that behaviour (e.g participating in an organizational

intervention) will lead to greater affective attachment to the organization. The second

strand posits that greater affective attachment to the organization will result in

behavioural outcomes. Consequently, organizational commitment may be viewed as an

outcome or an antecedent.

177

A number of studies investigating the impact of QCs on work outcomes have used

organizational commitment as an outcome (Bruning and Liverpool, 1993; Rafaeli,

1985; Griffin, 1988). With the exception of the last study, the remaining studies were

cross sectional in nature. Griffin (1988) in his study found no significant differences

between volunteers and non volunteers for the QC program prior to its introduction.

This would suggest that organizational commitment is not a differentiating factor

between those who volunteered for QC participation and those who did not. However,

Randall's (1990) meta analysis of organizational commitment-work outcome

relationships highlighted a weak positive relationship between organizational

commitment and effort. In addition, DeCotiis and Summers (1987) found that

organizational conunitment had a direct positive effect on employee motivation. Brief

and Motowidlo (1986) argue that components of organizational commitment indicate

dispositions toward prosocial behaviour directed toward the organization. This would

be consistent with previous research (Katz and Kahn, 1978) that committed individuals

are more likely to engage in extra role behaviours. Together these findings would

support the proposition that individuals who are committed to the organization are more

likely to exert effort on behalf of the organization. Following from this, it is

hypothesized that highly committed employees would be more likely to participate in

organizational interventions should the opportunity arise.

Higher order need strength is also hypothesized to have a positive impact on an

individual's participation in TQM. In other words, given the opportunity, individuals

with a greater need for achievement and satisfaction through work, are assumed to be

more likely to participate in TQM. Marks et al. (1986) found significant differences in

growth need strength between participants and non participants. The inclusion of higher

order need strength rests on a similar rationale to that provided by Hackman and

Oldham (1980) for job redesign. They hypothesize that individuals with strong needs

178

for growth are more likely to respond more positively to jobs high in motivating

potential than individuals with weak growth needs.

Higher order need strength has primarily been used as a moderator between work

redesign and the hypothesized outcome of intrinsic motivation. There is some debate

on the moderating effect of higher order need strength. Spector (1985) argues that six

published reviews of the area have failed to agree that a moderator effect exists. The

results of his meta analysis supports the Hackman and Oldham (1980) model with high

need strength individuals responding more predictably. Taking Rafaeli's (1985)

suggestion that higher order need strength may be one factor that differentiates

individuals reactions to QC activities, it is included in this model as a potential

predictor of employee participation in the intervention.

In view of the fact that QCs are designed to give employees greater participation in

their work and/or greater influence in decision making, it is hypothesized here that the

greater the difference between desired and perceived actual, the greater the likelihood

that an individual will participate in the intervention. Hill (1991b) found evidence that

QC members had a greater desire for participation in managerial decision making than

non participants. In contrast, Rafaeli (1985) found no difference in desired influence

between members and non members but found a significant difference in perceived

influence in favour of members.

Graham and Verma (1991) in the study of employee responses to employee

participation programs used a measure of the perceived gap between the desired and

actual influence. In their study, they found that the influence gap moderated the

positive relationship between affect toward Employee Participation Programs (EPPs)

and the length of involvement in the program. In other words, individuals who were

179

most dissatisfied with existing opportunities to participate in decision making had a

more positive view of the outcomes of EPPs than those individuals who were

reasonably satisfied with their opportunity to participate. Individuals who perceive a

gap between their desired level of influence and their actual influence will be more

likely to participate in the intervention in order to reduce this gap. In other words,

individuals who have less influence over decision making than they would like will be

more likely to participate in the intervention in the hope of increasing the influence they

have thereby reducing the gap between the influence they want and what they have.

This assumes, in the first instance, that individuals perceive the intervention as

providing the opportunity for greater influence over decision making.

Leader and participation Supervisor participative style is hypothesized to

positively affect subordinate participation. In terms of the classification adopted by

Cotton et al. (1988), supervisory participative behaviour would fall into the category of

informal participation. It is classified as such as it occurs through the interaction

between supervisors and subordinates. Steel and Lloyd (1988) note that supervisors

who encourage employee participation may feel more comfortable with QCs and be

more likely to choose the installation of QCs in their work area. In the context of

TQM, supervisory interaction with subordinates provides an opportunity for

subordinates to participate in the improvement process. Given the top down cascading

approach of TQM to affecting change, the involvement of employees is dependent upon

their immediate supervisor providing the opportunity for their participation. Klein

(1984) argues that the support of first line supervisors is a crucial factor if meaningful

changes are to take place in the workplace. Hence, in this model, it is hypothesized that

supervisory participative style will have a positive impact on employee participation in

the intervention.

180

8.3 Additional model and analysis procedures

This section presents a rationale for testing a variation of the model outlined in the

previous section. This is followed by a discussion of the analysis procedures adopted.

The second model includes a measure of perceived benefit of the intervention. The

rationale for including it in the model is that individuals who perceive the intervention

as providing some benefit to them will be more likely to participate in the intervention.

This is based on a strict assumption that individuals make a cognitive judgment prior to

participating in the intervention. However, it is quite possible that perception of the

benefit occurs as a result of participation. Also, it is plausible that perceptions of the

benefit may change over time. An individual may, for example, perceive the

intervention as providing benefit prior to the intervention but as a result of participation

may change his/her view of the benefit. This may occur as a result of high expectations

being set of the benefits the intervention will bring which an individual feels are not

met as a result of the experience of participation. The analysis is conducted exclusively

on employees at site 1 which implemented a TQM intervention between time 1 and

time 2.

All the independent variables included in the basic model were measured at time 1 prior

to the intervention. At this point, none of the respondents had any knowledge that the

intervention was going to take place thereby providing a rigorous test of whether

attitudes at time 1 can predict participation in the intervention at time 2.

The two models were tested separately using Ordinary Least Squares (OLS) regression.

The single new measure used in the analysis and not previously discussed is that of an

influence gap. This was measured by asking individuals how much influence they

would like to have over day to day work decisions that affect them. They were also

181

asked how much influence they had over day to day work decisions that affected them.

Their perceived actual influence was subtracted from their desired influence to give an

influence gap.

8.4 Descriptive statistics

Table 8.1 presents the results of t-tests between participants and non participants in the

intervention. As discussed in chapter 5, employees varied in the degree to which they

were participating in the activities of the intervention. To capture this variation,

individuals were asked the extent to which they were participating in the intervention.

Responses were elicited on a five point Likert scale (not at all to a very great extent).

Individuals who responded in the categories of "not at all" and "not much" were

treated as non participants. Individuals who responded in the categories of "to some

extent", "to a great extent" and "to a very great extent" were treated as participants.

Table 8.1: Independent t-tests between participants and non participants in the intervention.

Time 1

Participants (n=-86)

Non-participants (n=78)

Mean (S.D.) Mean (S.D.)

Organizational commitment 5.40 (0.90) 5.25 (1.06) Higher order need strength 6.00 (0.73) 5.74 (0.87)** Influence gap 0.80 (1.01) 0.88 (1.03) Supervisor participative style 5.35 (1.02) 4.58 (1.07)***

** = p < .05 ***=p < .01

As shown in Table 8.1, significant differences were found at time 1 (prior to the

intervention) between participants and non participants. These differences were found

182

in higher order need strength and perceived supervisor participative style with

participants being more positive. However, there were no significant differences

between the two groups in their commitment to the organization and the perceived gap

between desired and actual influence.

While the t-tests suggest that participants may have been more positive than non

participants on some dimensions at time 1, this is tested more fully and rigorously

below using regression analysis controlling for a series of demographic factors.

8.5 Results

Table 8.2 presents the results of the predictors of employee participation in TQM using

the predictors measured at time 1 and also including perceived benefit of the

intervention measured at time 2.

183

Table 8.2: Predictors of employee participation in a TQM intervention

Predictors (Time 1)

Time 2

Participation in TQM Participation in TQM

Job Tenure .07 .12 Age .09 .08 Gender -.09 -.09 Length of service -.164- -.14 Job Title 2 .11 .08 Job Title 3 .22*** .17** Supervisor participative style .30*** .23*** Organizational commitment .00 -.06 Higher order need strength .12 .08 Influence gap .13'- .11

Perceived benefit of TQM interventiont ;AP,2///0 //j /%;d4 .37***

Adjusted R2 .13 .25 N 164 164

+ = p<.1 ** = p<.05

*** = p<.01 t Measured at time 2

The results2 show that in the basic model, containing predictors measured at time 1

only, only one of the four hypothesized variables had a strong significant impact on

employee participation in the TQM intervention. This was supervisor participative

style. To a lesser extent (at 10% level), the gap between desired and perceived influence

was also found to have a positive impact on participation in the intervention. When

perceived benefit of the intervention is included in the model as a predictor of employee

participation, it was found to have a significant effect. The inclusion of perceived

benefit increases the explanatory power of the model from 13% to 25%.

2 The results of the analysis using logistic regression are presented in Appendix 15. Overall, the results yield similar results to those found using OLS regression.

184

8.6 Discussion

Supervisory participative style

The important finding is that supervisory participative style at time 1 is a significant

predictor of employee participation in the intervention at time 2. In other words, the

results indicate that supervisors who are participatively oriented prior to the

intervention are more likely to involve their subordinates in the TQM intervention.

This finding is consistent with the underlying approach to change underlying TQM;

that is, supervisors are the key mechanism for providing the opportunity for employees

to participate in TQM. It also suggests though, that the participative objective of the

intervention is more likely to be implemented by supervisors/managers who normally

manage along participative lines.

In this context, it is worth noting that employees who have participated in the

intervention do not perceive their immediate boss as becoming more participative as a

result of the intervention. Participants report a mean score of 5.35 at time 1 and 5.47 at

time 2 for supervisor participative style. This lack of significant positive change may

indicate that for this group of supervisors, the participative objective of the

intervention, in terms of, involving subordinates in the intervention, has been achieved.

What does not appear to have been achieved though is a move toward a more

participative style amongst this group of supervisors.

In contrast, for non participants, perceived participative style of their immediate boss

has increased from 4.58 at time 1 to 4.82 at time 2 (significant at 10% level). This may

indicate a move toward a more participative style of managing which needs to continue

so that subordinates are given the opportunity to participate in TQM. Time 2 represents

a point in the change process so in order to investigate whether, in the case of non

185

participants, greater supervisory participation leads to subordinate participation, it

would be necessary to monitor the process of change as it develops.

At least in the short term, the results indicate that supervisors who are generally more

participatively oriented will involve subordinates in TQM. This does not seem to be

due to an attitudinal or a behavioural change but simply because it is the normal style of

the supervisor. This issue is pursued in greater detail in a later chapter. For those

supervisors who are less participative in their style, in the short term, there does not

appear to have been a sufficient change in their participative orientation to allow

employee involvement in TQM. This is broadly consistent with Hill's (1991b) finding

where a number of managers failed to develop a participative oriented style of

managing to involve subordinates in the improvement process.

Organizational commitment

Organizational commitment is not a significant predictor of employee participation in

TQM. Participants and non participants were not significantly different at time 1 in

terms of their commitment to the organization. Consequently, the degree of

commitment to the organization does not differentiate participants from non

participants. In other words, individuals who are more committed to the organization

are not more likely to participate in the intervention than those who are less committed.

This is contrary to the hypothesis and is pursued later in the discussion. A

methodological point is worth discussing here regarding the use of cross sectional data.

If, for example, this analysis is restricted to the time 2 data; that is, replacing the

predictors measured at time 1 with the same predictors measured at time 2, the results

would indicate that organizational commitment has a significant positive effect on

participation in the intervention. From this, one would have argued that individuals

who are more committed to the organization would be more likely to participate in the

186

intervention than individuals who are less committed. However, the more rigorous

time 1 predictors as used here show that organizational commitment does not

significantly affect employee participation in the intervention.

Higher order need strength and influence gap

Higher order need strength and influence gap were not found to be significant in

predicting employee participation. Thus, individuals who have a greater need for

satisfaction and achievement through work are no more likely to participate in TQM.

Similarly, individuals who desire greater influence than they perceive they have are no

more likely to participate in TQM than individuals who have the influence they desire.

While participants and non participants are significantly different in terms of higher

order need strength at time 1, once other factors are controlled for, higher order need

strength does not have a significant effect on subsequent employee participation in the

intervention. This significant difference between participants and non participants at

time 1 goes against previous research (Marks et al., 1986). However, the results here

are in line with other research findings (Griffin, 1988). In terms of methodological

rigour, the results suggest that controlling for other factors, higher order need strength

or influence gap do not have a clear predictive effect on employee participation in the

intervention. This lack of significant effect may be due to the nature of employee

participation in TQM. This is discussed later as the explanation may be equally

applicable to the lack of significant effect of organizational commitment.

Perceived benefit of the intervention

Perceived benefit of the intervention, when added as a predictor of employee

participation in TQM, was found, (in addition to supervisor participative style) to have

a significant positive effect. In other words, individuals who perceive the intervention

as providing benefit to them in teims of their job are more likely to participate in TQM.

187

This assumes that individuals assess the intervention prior to participating in the

intervention and that perceived benefit is a predictor of participation. This may or may

not be the case. An individual may participate in the intervention and based on that

experience make an assessment of the intervention. That is to say, participation may be

a predictor of perceived benefit rather than the other way around. Furthermore, it is

plausible that individuals make an initial assessment based on the knowledge they have

at the time which may change in a positive or negative direction when they have

acquired further information or participated in the intervention. As employees'

participation in the intervention and their assessment of it was collected at the same

point in time, it is difficult to discern the sequence of events in relation to participation

and assessment. The subsequent chapter picks this up when it looks at the antecedents

of assessment of the intervention.

In summary, the findings indicate that the only clear and significant predictor of

employee participation in the intervention is the participative orientation of their

immediate supervisor. Consequently, in the short term, employee participation will

take root where employees are accustomed to a more participative style of managing.

The results highlight the distinctive nature of employee participation in TQM compared

to other types of employee participation programs (EPPs). The primary difference

relates to the participative mechanism. In EPPs, a mechanism is put in place

independent of employees' supervisor and employees are asked to volunteer for

participation in the program. Consequently, employee participation is largely

independent (or not dependent upon) of the actions of the supervisor. Given the

cascading approach to change inherent in TQM, employee participation is dependent

upon the supervisor allowing and facilitating their involvement. In other words,

regardless of how committed an individual is to the organization or how much more

influence they desire, they cannot participate in the intervention unless they are given

188

the opportunity to do so and this opportunity occurs via the behaviour of the immediate

boss. Therefore, while TQM has the potential to overcome the problems of QCs (Hill,

1991b), it is faced with a potentially different obstacle; that is, having to change the

style of managing of supervisors to allow employee participation.

Graham and Velma (1991) conclude that in order to ensure that employees view the

outcomes of EPPs favourably, it is necessary to manage individual "proximity" to

EPPs. In other words, it is important to plan rapid diffusion of the EPP so as to

increase individual proximity to the program. Chapter 5 highlighted that non

participants in the intervention experienced a reduction in their team orientation. The

explanation offered was that this may be due to unmet expectations as a result of their

lack of involvement. Consequently, attention was directed to a more rapid diffusion of

the opportunity for employees to participate in the intervention. The results of this

chapter highlight that supervisory behaviour may present a stumbling block for the

rapid diffusion of employee participation in a TQM intervention.

8.7 Conclusions

To conclude, restricting the analysis to using attitudes and perceptions of behaviour

measured at time 1, the only significant predictor of employee participation in TQM is

supervisory participative behaviour. This is consistent with TQM and its cascading

approach to change. Unlike QCs, TQM places the responsibility for employee

participation firmly in the hands of managers and supervisors.

Placing the responsibility for employee participation in the hands of supervisors may be

a double edged sword. It overcomes the problem of QCs by integrating supervisors

into the improvement process but it also has potential damaging consequences if

189

supervisors are reluctant to take on board the principles and practices of the TQM

intervention. What this chapter highlights is the importance of supervisory behaviour

pre intervention in affecting employee participation in the intervention. The subsequent

chapter complements this analysis by looking at the antecedents of employee

assessment of the intervention.

190

Chapter 9: Predictors of Employee Assessment of a TQM Intervention

9.1 Intro duction

The previous chapter showed that the most important predictor of employee

participation in TQM is supervisory participative style. In other words, supervisors

who are participatively oriented in their style of managing prior to the intervention are

more likely to involve their subordinates in the TQM intervention. However,

participants in the intervention did not perceive their immediate supervisor as becoming

significantly more participative between time 1 and time 2. Thus, the intervention has

not significantly altered the style of managing for those supervisors who generally

operate along participative oriented lines, rather, it has facilitated the cascading of the

intervention to the lowest level in the organization.

This chapter continues this line of investigation by looking at the predictors of how

individuals assess the intervention. From the previous chapters, the importance of

assessment has been highlighted as having a significant impact. Overall, how

individuals perceived the intervention was more significant in affecting attitudes than

participation in the intervention per se. Given the importance of how an individual

perceives and assesses the intervention, this chapter sets out to explore the antecedents,

if any, of how individuals assess this particular organizational intervention. Two

related but distinct questions are addressed: first, are there any attitudes that predispose

individuals to respond positively to organizational interventions, in this case, TQM?

Second, are the predictors of assessment the same as the predictors of participation?

191

Returning to the research on Quality Circles (QCs), considerable emphasis is placed on

participation as the key mechanism in affecting subsequent attitudinal and performance

outcomes. Rafaeli (1985) investigates the relationship between QC membership and

outcomes such as job satisfaction, influence and job characteristics such as variety and

autonomy. While the author does not include an assessment or evaluation of QC

activities as an explanatory variable in linking membership to outcomes, there is a

suggestion that some employees will react more positively to QC activities. Griffin's

(1988) longitudinal analysis of the link between participation in QCs and job

satisfaction, organizational commitment and performance outcomes found that there

was an improvement in outcomes to the 18 month mark and a subsequent decrease.

The author presents a possible explanation in that an increasingly mechanistic approach

was adopted to and there was a decline in interest and enthusiasm for QCs.

A number of studies found that membership in QCs is related to perceived participation

levels (Marks et al., 1986; Steel, Mento, Dilla, Ovalle, and Lloyd, 1985; Bruning and

Liverpool, 1993). In one study, expected changes in perceived participation occurred in

one group but not the other (Steel et al., 1982). A potential explanation put forward by

the authors is that the group that did not change regarding perceived participation had

not had enough time to develop. Therefore, the development and implementation of the

QC process may be an important explanatory factor in the participation-outcome link.

Also, management attitudes toward QCs could affect an individual's experience of QC

activities (Bruning and Liverpool, 1993). It is widely accepted that individual

differences are important moderators of how individuals respond to organizational

stimuli (Staw, Bell and Clausen, 1986). Therefore, it is possible that differences may

exist between participants in QCs in terms of how they assess or evaluate QCs.

192

What may be important in explaining the outcomes of QC activities is an individual's

assessment of those activities. Steel and Jennings (1992) note that where no attitudinal

change was found as a result of a QC intervention, this may be due to unmet

expectations of participation. There may be other factors that may act as an inhibitor of

attitudinal change resulting from QC participation, for example, lack of supervisory

support for QCs or a lack of perceived benefit of participation.

It would seem that much of the research assumes that participation in QC activities is

viewed in a positive manner by its participants. Relatedly, individuals who volunteer to

participate in QC activities, prior to their participation, are more likely to view QC

activities positively. The voluntary nature of membership in QCs has been widely

documented (Cole, 1980). However, there has been little systematic investigation into

the real extent of voluntary participation. Lillrank and Kano (1989) in their study of

how QCs operate in Japan, highlight the use of informal pressure to ensure

participation. Similarly, McArdle et al. (1995) note the use of covert encouragement to

participate in QCs as part of TQM in their case study. The notions of voluntary

participation and participation as a positive experience are very much interrelated.

First, individuals who volunteer are more likely to view participation as positive.

Second, individuals who participate may be disappointed due to unmet expectations

and consequently withdraw their membership. Thus, the assumption of participation as

a positive experience may only hold true if in fact participation is based on a true

voluntary basis allowing individuals to withdraw as easily as they volunteered. In

organizational settings, informal pressure may be placed on an individual by colleagues

or by the immediate boss to participate and continue to participate. In this situation, an

individual's assessment of QC activities may be more important that their participation

in predicting subsequent attitudinal and performance changes.

193

Graham and Verma (1991) in their study set out to explain an individual's assessment

of an employee participation program using a combination of dispositional and

situational predictors. They propose that an individual's experience may or may not be

satisfactory due to individual differences. Also, they argue that participation will be

assessed positively only if the experience is a positive one. In their study, they do not

attempt to link an individual's assessment of the program to subsequent outcomes such

as job satisfaction and organizational commitment that are widely used in the QC

studies and in empirical investigations of participation. Steel et al. (1985) in their study

of the factors that influenced the success and failure of two QC programs found

significant differences in participant assessment of QC activities between the successful

and unsuccessful programs. Success was interpreted as having a positive impact on a

range of attitudinal outcomes. Participants in the successful QC program reported

higher managerial support for QC activities and greater satisfaction with the QC

process than participants in the unsuccessful QC program.

Research on training has highlighted the importance of reactions to training activities in

influencing posttraining attitudes (Tannenbaum and Yukl, 1992; Noe and Schmitt,

1986; Noe, 1986). Kirkpatrick's (1976) hierarchical model of training outcomes

suggests that four outcomes need to be considered in evaluating training programs;

trainees' reactions to the program, learning, behavioural change and individual and

organizational performance. The model hypothesizes that each training outcome affects

the next level in the hierarchy, for example, trainee reactions will have an important

influence on learning and so forth. A number of training evaluation studies have

provided some support for the hierarchical model (Latham, Wexley, and Pursell, 1975).

Noe (1986) states that the strongest evidence is provided by Clement (1978) where

trainee reactions had a causal impact on learning and learning had a significant

influence on behaviour change. Tannenbaum and Yukl (1992) argue that individual

194

differences in attitudes and expectations may have a central role in influencing

posttraining attitudes. In their study, they found support for the influence of training

fulfillment, reactions and perfoiniance on posttraining attitudes. Training reactions

were positively related to posttraining motivation and commitment (Tannenbaum and

Yukl, 1992).

The empirical research on training suggests that it is not participation per se in the

training activities that influence subsequent attitudes, the emphasis is on trainee

assessment in terms of expectations, desires and reactions as having a greater impact on

posttraining attitudes. Applying this to employee participation programs, an

individual's assessment of the program may have a greater impact on subsequent

attitudes than their participation per se. Consequently, rather than assuming that

participation in, for example, a QC is a positive one based on voluntarism, it may be

more worthwhile as in the study of Graham and Verma (1991) and Steel et al. (1985)

to test this proposition.

This notion of employee assessment is highlighted quite recently by Marchington et al.

(1994). At a broad level, they argue (in the context of employee involvement) that

while employee involvement schemes are directed at employees, there is a noticeable

lack of empirical research investigating employee attitudes and experiences of such

schemes. More specifically, two points are made. First, the authors argue that undue

emphasis is placed on individual demographic characteristics (age, gender, length of

service) as a way of differentiating between individual response to these schemes. This

is to the neglect of for example, the competitive environment or managerial style.

Second, knowing if employees like or dislike the scheme is irrelevant unless one can

ascertain "why people hold those views, and what factors cause them to vary" (p891).

This chapter addresses to some extent the first point in that, not only are demographic

195

characteristics used but also employee attitudes on a range of dimensions (for example,

organizational commitment) to predict employee responses. In view of the second

issue raised, this study, in the specific context of TQM, investigates the factors which

influence or predispose individuals to assessing the intervention in more favourable or

unfavourable terms.

This chapter continues by outlining two models to be tested and their rationales. The

first model replicates the model used in the previous chapter to predict employee

participation in TQM. The second model includes an individual's prior experience or

assessment of a participation program as a predictor of how individuals assess the TQM

intervention. In addition, an individual's participation in the TQM intervention is also

included as a predictor of assessment. Subsequent to presenting the models, the

measures and analysis procedures are briefly presented, followed by a discussion of the

results.

9.2 An hypothesized model of employee assessment of a TQM intervention

Two hypothesized models of employee assessment of the intervention are tested. The

first mirrors the model outlined in the previous chapter for employee participation.

This will allow a direct comparison of the predictors of participation in and assessment

of the intervention. The variables include supervisor participative style, organizational

commitment, higher order need strength and the gap between desired and perceived

actual influence. These variables are hypothesized to have a positive effect on how an

individual assesses the TQM intervention.

196

The second model (presented in Figure 9.1) includes two additional variables. The first

additional variable taps an employee's previous experience or assessment of a

participation program similar to what is widely known as QCs. The rationale for

including this previous experience in the model is that an individual's prior experience

with organizational interventions may have an impact on their assessment of future

interventions. For example, an individual who participated in the previous QC program

and who was very satisfied with its operation may be more likely to assess future

participative interventions in a more positive manner

At any given point in the implementation of a TQM intervention, there will be variation

between employees as to their extent of participation in the intervention. From the

previous chapter, it was shown that employees with a participative supervisor became

involved more quickly and to a greater extent than employees who had a supervisor

who was less participatively oriented. There will be employees who have decided that

they will not participate or who are reluctant to do so and employees who are willing to

participate but who have not yet had the opportunity to participate. Graham and Verma

(1991) in their cross sectional study of the predictors of employee responses to

employee participation programs (EPPs) found support for their hypotheses that the

closer an individual is to membership and the longer their involvement in EPPs, the

more positive their attitudes about EPPs. From this, one could hypothesize that the

greater an individual's participation in the intervention, the more positive their

assessment. Following from this, an individual's participation in the intervention is

included in the model to investigate whether the degree of participation has an impact

on the assessment of the intervention.

197

O

Perc

eive

d be

nefit

of t

he

TQM

inte

rven

t ion

Influ

ence

gap

F. 4-4

4-0 E

198

9.3 Measures

All the measures used in the subsequent analysis have been discussed in previous

chapters with one exception. This measure taps an employee's previous experience or

assessment of a participation program. It is the creation of this measure that is now

discussed.

Change interventions do not occur in an organizational vacuum and in approaching

these interventions, individuals bring with them their previous experience (positive or

negative) which may influence how they assess future interventions. As noted in

chapter 2, at the commencement of this study, a small number of individuals were

participating in a Continuous Improvement Group but overall, this initiative was

rapidly deteriorating in terms of managerial and employee support. Consequently,

individuals were broadly categorized as falling into one of three categories: presently

participating, would like to participate and having no desire to participate in a

Continuous Improvement Group. As discussed below, this three tier categorization was

extended to five so as to include how individuals assessed their present participation or

expectations of future participation.

Individuals were asked if they were presently participating in a Continuous

Improvement Group (this is similar to QCs in that employees volunteer to participate

and meet in work time to discuss work problems and suggest improvements). If

individuals responded that they were participants, they were then asked to respond to a

series of questions tapping their satisfaction with managerial support, recognition for

improvements and the extent to which their participation was making use of their

abilities and helping them in their work. This group was subdivided into those who

199

viewed their participation in a very positive manner and those who viewed it less

positively.

The remaining individuals were asked if they would like to participate in a Continuous

Improvement Group. Those who replied that they would were subdivided into two

groups based on their replies to three questions tapping their anticipated benefit of

participation in terms of helping them in their work, making use of their abilities and

knowledge and the degree it would help them make improvements in their work area.

Respondents were divided into two categories based on their perceived expectations;

high expectations of the benefits of participation and lower expectations. Individuals

who responded that they had no desire to participate in a Continuous Improvement

Group were categorized as one group. Five groups were created based on an

individual's prior experience or assessment of this participative program. The groups

are as follows:

Groups: previous experience/ assessment at time 1

1 - Not participating and has no desire to participate

2 - Not participating but has a desire to participate with low expectations of the benefits

3 - Not participating but has a desire to participate with high expectations of the benefits

4 - Participating but which is not viewed very positively

5 - Participating which is viewed positively

200


The two models were tested separately using OLS regression and controlling for the

usual set of demographic factors. For the first model, the variables included were the

same as in the model predicting participation in the intervention. As previously noted,

in the second model, two additional predictors were included. The first additional

predictor was an individual's prior assessment or experience of participative programs.

Dummy variables were created that correspond to the groups outlined in the previous

section. The reference category used was group 1; those individuals who at time 1 were

not participating and had no desire to participate in Continuous Improvement Groups.

Present participation in the TQM intervention was also included in this model. This

predictor differs from the remaining predictors in that it was measured at time 2 while

the others were measured at time 1. Similar to the previous chapter, the analysis is

conducted on the sample of employees at site 1.

9.5 Results

Table 9.1 presents the results of the initial model that mirrors the model tested in the

previous chapter on employee participation in the TQM intervention.

201

Table 9.1: Predictors of employee assessment of a TQM intervention (using same predictors as for employee participation)

Predictors (Time 1)

TIME 2

Perceived benefit of TQM intervention

Perceived appropriateness of TQM

intervention

Job Tenure -.13 -.12 Age .03 .06 Gender -.01 .03 Length of service -.03 -.03 Job Title 2 .09 -.15** Job Title 3 .13 -.03 Supervisor participative style .19** .30*** Organizational commitment .17** .30*** Higher order need strength .11 .14'. Influence gap .03 .06

Adjusted R2 .09 .26 N 166 166

+ = p<.1 ** = p<.05 *** = p<.01

The results show that supervisor participative style and organizational commitment

prior to the intervention are significant predictors of perceived benefit and

appropriateness of the intervention. In addition, higher order need strength was found

to be significant (at 10% level) in predicting an individual's assessment of the

appropriateness of the intervention. The influence gap; that is, the discrepancy between

desired and perceived actual influence, was not found to significantly affect how

individuals assessed the intervention.

202

In comparing the predictors of participation with the predictors of assessment, there is

only one common predictor, that of supervisory participative style. Organizational

commitment was not found to significantly affect an individual's participation in the

intervention but it had a significant effect on how individuals assessed the intervention.

Table 9.2 presents the results of the full model that includes present participation in the

intervention and previous assessment of a participative program.

Table 9.2: Predictors of employee assessment of a TQM intervention (full model)

Predictors (Time 1)

TIME 2

Perceived benefit of TQM intervention

Perceived appropriateness of TQM

intervention

Job Tenure -.14+ -.14** Age .02 .04 Gender .05 .09 Length of service .06 .06 Job Title 2 .02 -.22** Job Title 3 .01 -.15** Supervisor participative style .04 .18** Organizational commitment .13 1- .26*** Higher order need strength .04 .09 Influence gap -.02 .01 Participation in TQMt .38*** . 33*** Previous participation 2 .12 .17** Previous participation 3 .13 .15** Previous participation 4 .10 .14** Previous participation 5 .16** .10

Adjusted R2 .22 .37 N 164 164

+ = p<.1 ** = p<.05 t measured at time 2

203

Participation in the intervention has a significant effect on assessment of the

intervention. In other words, the more an individual participates in the intervention, the

more positive the assessment. Organizational commitment continues to be significant

in affecting how individuals assess the intervention. Previous experience was also

found to have a significant effect. In the case of perceived benefit, group 5 are more

likely to view the intervention as providing benefit. This group held positive views on

their previous experience in a participation program. In addition, previous experience

played a significant role in affecting whether individuals deemed the intervention as

appropriate.

9.6 Discussion

Predictors of participation vs. assessment

One of the significant findings is that the predictors of assessment do not mirror the

predictors of participation discussed in the previous chapter. The difference is not

surprising given the nature of participation versus assessment; that is, behaviour versus

attitude. As discussed in the previous chapter, employee participation is dependent

upon the actions and behaviour of the supervisor. Thus, unless a supervisor involves

his/her subordinates in the intervention, there is no alternative method for employees to

participate. While employee behaviour (i.e. participation) is dependent upon the

behaviour of the supervisor, employee attitudes (toward the intervention) are influenced

not only by the behaviour of the supervisor but also by other attitudes held by

employees.

This result highlights the difference between QCs and TQM. QCs rely on voluntary

participation from employees and give a minority role to supervisors; that is,

204

supervisors are given the responsibility to oversee QC activities but are not allocated

parallel authority. Consequently, QCs can operate outside the sphere of supervisory

influence in determining their own agenda without being required to consider the

priorities of the supervisor. This limited role allocated to the supervisors has been

previously highlighted as a stumbling block to the QC grass roots approach to change

(Collard and Dale, 1989; Hill, 1991b). QCs in Japan are firmly integrated into TQM

and are considered part of an organization-wide effort of quality improvement and

change (Lillrank and Kano, 1989). Thus, in theory, TQM overcomes the limited role

assigned to supervisors in QC activities. Supervisors are allocated the role of cascading

the principles and practices of TQM and of integrating quality improvement as a core

activity of the organization.

TQM places greater emphasis on including all employees (Lawler, 1994) with the

notion of "total involvement" (Oakland, 1989). This raises the question of the nature of

voluntary participation of employees in TQM. One could argue that as TQM places

supervisors in a crucial position regarding implementing and sustaining quality

improvements at the lower levels in the organization and in conjunction with the

proximity of the supervisor to employees, it may be more difficult to operate employee

participation on a voluntary basis as advocated by numerous writers (Juran, 1989;

Ishikawa, 1985). The vested interest of the supervisor in succeeding (for whatever

reason, the avoidance of sanctions or the pursuance of rewards) may lead to informal

pressure being placed on subordinates to participate; voluntary participation being

espoused but not practised. This may very well be a contributing factor to the lack of

impact of participation on the core outcomes of TQM and also highlight why, in this

case, assessment is of greater importance. Thus, while TQM overcomes the

"problematic" role of supervisors in QCs, it raises the issue of voluntary participation

205

of employees but moreso, it places the responsibility of quality improvement at the

lowest level firmly on the supervisor.


In comparing the predictors of participation in (chapter 8) and assessment of the

intervention, organizational commitment was found to have a significant effect on an

individual's assessment but not their participation. What the results indicate is that

individuals who are more committed to the organization prior to the intervention are

more likely to assess the intervention as being appropriate and as providing benefit. In

addition, organizational commitment seems to have a dual role in being an antecedent

and an outcome of the intervention.

An individual's identification with the organization and willingness to exert effort on

behalf of the organization may positively colour an individual's assessment of

organizational activities. The previous chapter showed that commitment levels of

participants significantly increased while the level of non participants remained stable.

This chapter has shown that organizational commitment is an antecedent of how

individuals assess the TQM intervention. This is consistent with the proposition put

forward by Eisenberger et al. (1990) that affective attachment to the organization would

create "evaluation biases in judging the organization's actions and characteristics"

(p57).

Treating organizational commitment as an antecedent and an outcome is consistent with

previous research on training (Gist, 1987; Latham, 1989; Tannenbaum and Yukl, 1992).

Tannenbaum and Yukl (1992) argue that organizational commitment is likely to

influence whether an individual views the training as useful to themselves and the

organization. In terms of organizational commitment as an outcome, employees may

206

view training as representing the willingness of the organization to invest in them, and

reciprocate by increasing their affective attachment to the organization.

A similar explanation is put forward to account for the role of organizational

commitment as an antecedent and an outcome of the TQM intervention. If TQM is

viewed as a way to ensure the future prosperity of the organization and is presented as

something an organization needs to do, individuals who are highly committed to the

organization may be more likely to view the intervention as providing benefit; ensuring

the future success of the organization and as a consequence providing an economic

benefit for the individual of continued employment. Alternatively, or in addition, an

individual who is strongly committed to the organization may be more likely to identify

with the core values of the organization. If a particular organizational intervention is

broadly consistent or does not represent a radical departure with those values, an

individual may be more likely to view the intervention in positive terms. Using

cognitive self concept theories, Reger et al. (1994) address why beneficial change is

often resisted by loyal members who want what is best for the organization. This

perspective suggests that organizational initiatives which radically depart from the

organization's past fail due to members' cognitive structures which constrain their

understanding and support for new initiatives.

Higher order need strength

It was hypothesized that an individual's higher order need strength would have a

positive impact on how the intervention was assessed. Although the beta coefficients

are quite high (significant at the 10% level for perceived appropriateness ), the effect of

higher order need strength was not found to have a strong effect. As discussed in the

previous chapter, this finding goes against some previous research (Marks et al., 1986)

but is also consistent with other findings (Griffin, 1988).

207

Influence gap

The influence gap did not have a significant impact on perceived benefit or

appropriateness of the intervention. Individuals who have a greater discrepancy

between their desired and actual influence prior to the intervention did not perceive the

intervention as more appropriate or beneficial. This is contrary to the findings of

Graham and Verma (1991) where individuals reporting the highest affect level towards

EPPs were individuals who were inactive organizational citizens or who perceived the

greatest participation gap. Two possible explanation are put forward for the lack of

significant effect of influence gap on perceived benefit and appropriateness of the

intervention. First, higher order need strength may be suppressing the effect of the

influence gap. One could plausibly argue that individuals with strong growth needs

would desire the opportunity for greater influence in order to satisfy a need for

achievement through work. Second, the measure of influence gap is a single item

measure unlike Graham and Verma's (1991) twelve item measure. Consequently, the

lack of effect may be due to the limited scope of the measure.

Prior experience of a participative program

Up to now, the discussion has focused on the difference between the predictors of

participation in and assessment of the intervention and why this may be the case. The

second model contained two additional predictors that are worth discussing

commencing with an individual's prior experience of a participative program. This

predictor is important for two reasons. First, individuals carry with them prior

experiences of organizational activities which influence their interpretation of future

events. Second, it has been documented that QC type participative programs are

transitional mechanisms (Lawler and Mohrman, 1985) in the search for a more effective

way of gaining quality improvements. Hill (1991b) reports that of the thirteen

organizations that experimented with QCs, seven continued to make provisions for

208

participative quality improvement primarily through the adoption of TQM.

Consequently, this raises an issue of whether experimentation with a particular

participative mechanism at one point has an effect on subsequent participative

interventions in terms of employee assessment.

The results show that previous experience or assessment does have an impact on

perceived benefit and appropriateness of the intervention. Individuals who had prior

positive experience with Continuous Improvement Groups are more likely to assess the

intervention in positive terms regarding perceived benefit. In predicting perceived

appropriateness of the intervention, compared to the group who reported no desire to

participate at time 1, the remaining groups (except those who had a positive experience

at time 1) viewed the intervention as being more appropriate. Thus, previous

experience or assessment of a participative program seems to influence how individuals

assess future interventions. Certainly, in the short term, it would seem that individuals

do carry with them their previous experience which has some influence on how they

assess future organizational interventions. This is consistent with the argument

presented by Marchington et al. (1994) who suggest among other things, that

employees attitudes to employee involvement programs are dependent on their past

experience of such schemes. However, whether this holds true in the longer term is

another question.

Participation in the intervention

A substantial number of studies have investigated the link between participation in QCs

and various attitudinal and in fewer cases performance outcomes (Griffin, 1988;

Rafaeli, 1985; Marks et al., 1986; Head et al., 1986). The emphasis on the direct

participation-outcome linkage has been to the neglect of an assessment of the

participation activity, in this case, QCs. Very few studies have adopted or tested a

209

participation-assessment-outcome linkage. Graham and Verma (1991) in their study of

EPPs and the affect toward them found that proximity to and duration of involvement

in an active EPP were strongly associated with EPP affect. As such they are primarily

testing the participation-assessment linkage In the context of this study, several

questions are raised. Are there factors, other than participation, that affect an

individual's assessment of the intervention? Does an individual's participation in the

intervention eliminate the effects of previous experience in their assessment of the

intervention?

Participation in the intervention was found to have a significant positive effect on the

two assessment variables. Thus, the more an individual participates in the intervention,

the greater the perceived benefit and appropriateness of the intervention. This finding

is broadly consistent with that of Graham and Verma's (1991) but goes further in that

the greater the participation, the more positive the assessment. However, as

respondents' degree of participation was measured at the same time as their assessment

of the intervention, it is not possible to ascertain whether participation leads to

assessment or whether the reverse relationship holds true. It is possible that

participation and assessment are interrelated; that is, an individual, based on

information about the aims and purpose of the intervention, may initially make an

assessment which may subsequently alter based on his/her involvement in the

intervention. Indeed, it is quite plausible that individual assessment of the intervention

is not a stable characteristic and may alter depending on how the intervention develops

and how the individual reacts to this.

Thus, while participation has a significant impact on how individuals assessed the

intervention, it is not the sole predictor. From a methodological stance, it is important

to remember that participation was measured at time 2. Consequently, from this, one

210

could argue that its effect may be inflated in relation to the remaining predictors

measured at time I. However, other attitudes prior to the intervention also have a

significant effect; an individual's attitude toward the organization, perception of

supervisory behaviour and previous assessment of a participative program.

To summarise, the findings from this and the previous chapter highlight several issues.

First, the predictors of participation in and assessment of the intervention are not the

same. Employee participation in the intervention is predicated on the behaviour of the

supervisor. In other words, the intervention has cascaded where the supervisor(s) have

a participative style of managing and not because there has been a significant shift in

the participative style of the supervisors. In contrast, employee assessment is also

influenced by their commitment to the organization prior to the intervention. Second, a

potential explanation put forward for the lack of predictive power of participation on

subsequent attitudes may be the result of informal pressure being applied to employees

to participate. Due to the proximity of supervisors to employees and their perceived

authority, employee participation may be subject to some degree of informal pressure.

Finally, while participation in the intervention has a significant positive effect on

assessment, it is not the sole significant influence. Rather, employee assessment of the

intervention is influenced by a range of diverse factors to include an individual's prior

experience and assessment of a participative program.

9.7 Conclusions

This chapter set out to investigate the predictors of assessment of the TQM

intervention. Predictors of participation in and assessment of the intervention were

found to be different. They did share a common predictor, that of supervisor

participative style prior to the intervention. The results indicated that organizational

211

commitment prior to the intervention had a significant positive effect on how

individuals assessed the intervention.

The difference in the predictors between employee participation and assessment may

shed light on why employee participation does not have a significant effect on any of

the TQM outcomes measured here. As discussed in chapter 8, employee participation

is dependent upon the actions of the supervisor. What is unknown is the degree to

which informal pressure is placed on the individual by his/her immediate boss or

colleagues to participate. Therefore, participation in the intervention may not be based

on a true voluntary basis. If this is the case, the assumption that participation is a

positive experience may not hold true. Consequently, how employees assess or judge

the intervention is more important in affecting attitudinal and behavioural change.

Present participation in the intervention did have a significant effect on how the

intervention was assessed. However, it was not the only significant influence. An

individual's prior experience of participative programs was found to have a significant

effect on their assessment of the present intervention.

Up to this point, the focus of investigation has been at the level of employees. The

subsequent chapter takes the analysis a step further by concentrating on

supervisors/managers.

212

Chapter 10: Supervisory Behaviour

10.1 Introduction

The emphasis of the thesis so far has been on employees. Two lines of investigation

were pursued: evaluating the impact of the TQM intervention on employee attitudes

and; predicting employee participation in and reaction to the intervention. Throughout

this investigation, supervisory behaviour was highlighted as having an important

influence in affecting employee attitudes. Supervisor participative style and supervisor

commitment to quality respectively, were found to have a significant effect on team

orientation and commitment to improvement. Subsequently, it was shown that the

participative style of the supervisor was an important determinant of employee

participation in the TQM intervention. In a similar vein, supervisor participative style

was highlighted as having some influence on how employees reacted to the

intervention. Up to this point, the thesis has not addressed supervisory behaviour. In

light of the importance of supervisory attitudes and behaviour in affecting employee

attitudes and behaviour, this chapter examines the predictors of supervisory

participative style and commitment to quality.

The importance of supervisory behaviour has long been recognized as having a major

influence in organizational life. An abundance of empirical research exists detailing the

impact and importance of the behaviour of those in the higher echelons on a range of

organizational characteristics, processes and outcomes. The behaviour of supervisors

can be thought of, in broad terms, of having an effect on the behaviour and attitudes of

employees in their everyday activities and also play a pivotal role in an organization's

attempt to change. The work of classic human relations theorists asserted that

leadership assumptions of managers and their consequent processes were primary

213

determinants of climate which in turn, affected individual behaviour (Likert, 1967;

McGregor, 1960). Further developments on the impact of various leadership styles on

employee behaviour have been widely researched and documented.

The emphasis of human relations theorists is similar to that of transactional leadership

in focusing on leadership at the lower hierarchical levels. In contrast, a more recent

strand of the leadership research has focused on transformational leadership which

emphasizes the communication of values and the implementation of a vision.

Transformational leadership processes have the potential to enhance followers' work

oriented values to be congruent with those of the group or organization (Burns, 1978;

Conger and Kanungo, 1987). Waldman (1994) argues that several processes are likely

to operate; one process is a clear and appealing vision. In addition, the leader acts as a

role model for the espoused values. This latter process, leading by example, is

consistent with the "Pygmalion effect" (Livingston, 1969; Eden, 1984). This effect

relies on the role expectations of the leader affecting the behaviour of followers. In

other words, a leader's expectations of subordinate behaviour is communicated to them

through the behaviour of the supervisor and is hypothesized to affect subordinate

behaviour. Scott and Bruce (1994) provide support for the Pygmalion effect in the

context of innovation; that is, the role expectations of a supervisor in relation to

subordinate innovation had an effect on the innovative behaviour of subordinates.

In addition to the effect of role modeling, it has been asserted that the supervisor is a

potentially powerful source of social information (Griffin, 1983). In the context of task

attributes, Griffin (1983) found that employee perceptions of task attributes and

affective responses were significantly influenced by both the objective task changes and

the informational cues provided by the supervisor. Thus, supervisors may affect

employee attitudes through their own behaviour and through the informational cues

214

they provide as to organizational processes and events. In investigating supervisory

attitudes toward employee involvement programs, Klein (1984) found that supervisors

rarely exhibited overt resistance to top management initiatives. Occasionally,

supervisors criticized the programs with peers and subordinates but more often, they

remain silent or communicate mild enthusiasm. This is interpreted by subordinates as a

lack of support for the program.

In relation to organizational change programs, individuals in managerial positions have

an important role to play in affecting change. Whether it be cultural change or changes

in the structuring of work, commitment from management is a vital ingredient for

successful change efforts (Schein, 1985; Walton, 1977). It has also been asserted that

the commitment of supervisors is essential if meaningful changes are to occur in the

workplace (Klein, 1984). The discussion so far, at a general level, has highlighted the

importance of those in supervisory positions in affecting the attitudes, behaviour and

perceptions of employees. This is very much applicable in the context of TQM. The

objective of TQM is to change the attitudes and behaviour of management and

subsequently, the attitudes and behaviour of employees. In terms of process of change,

the initial focus is on changing individual attitudes and behaviour within the managerial

hierarchy and using this change as a mechanism for changing the attitudes and

behaviour of employees. The prescription from the TQM proponents is quite clear, a

change in management attitudes and behaviour will lead to a change at employee levels

and a lack of change at managerial levels will presumably lead to a lack of change at

employee levels.

Management commitment to TQM has been widely documented and espoused as a

necessary ingredient for the success of TQM (Deming, 1986; Juran, 1989; Hill, 1995).

Hill (1995) argues that top management is the key champion of TQM within the

215

organization reflecting TQM's strategic importance. They set the quality priorities,

establish and facilitate an appropriate culture, provide resources and lead by example.

Waldman (1994) argues that the concept of transformational leadership appears to be

consistent with the prescription for managers from the quality proponents (Deming,

1986; Juran, 1989; S ashkin and Kiser, 1993). He specifically asserts that

transformational leadership may be the mechanism by which managers can affect

individual values in order to achieve teamwork and continuous improvement. Thus, the

onus is on management to change and provide role models for other organizational

members to follow.

In addition, TQM requires a change in managerial style. As Wilkinson (1994)

indicates, there are common underlying themes between HRM and TQM to include a

view of employees aligned to McGregor's Theory Y. Wilkinson and Willmott (1995)

note that TQM requires a shift from Theory X to Theory Y, in the pursuit of achieving

quality (Willmott, 1992;1993) and not for individual self actualization a la McGregor

(although this may occur in the process of achieving quality, it is not the objective). Hill

(1991b) argues that the task for top management is to create a culture that facilitates

participation of all employees in the pursuit of continuous improvement. Underlying

this participation is a cascading effect; middle managers and supervisors are allowed

greater influence over the decisions that affect them and they, in turn, adopt a more

participative approach thus providing a climate for employees to participate in the

improvement process.

In terms of implementation, TQM starts at the top of the organization and cascades

down the organization. Consequently, the first line supervisor / manager plays a critical

role in instilling TQM values at employee levels. Moreso, as the supervisors are the

216

salient representative of management in the eyes of those at the bottom of the hierarchy,

they play a key role in setting the quality agenda at the bottom of the organization.

Given the centrality of change in managerial attitudes and behaviour to the TQM

process, the first issue that warrants examination is the change, if any, that has occurred

within the entire group of managers / supervisors at the site (n=51). Subsequently, the

chapter sets out to investigate the predictors of supervisory participative style and

commitment to quality as these have been portrayed as two key dimensions of

supervisory behaviour in the context of TQM. The focus here is on first line

supervisors / managers (n=27). As previous chapters have shown, it is an employee's

immediate supervisor that is the most salient in affecting employee attitudes and

behaviour.

This chapter begins by looking at the change that has occurred between time 1 and time

2 in the entire group of managers / supervisors. This group has been directly involved

in the intervention and therefore it was not possible to establish a post hoc control

group. This provides the context in which the predictors of the two dimensions of

behaviour of first line managers / supervisors may be examined. The rationale for

concentrating on first line managers / supervisors is that they are more proximal to

employees. To do this, an hypothesized model of supervisor participative style and

commitment to quality is outlined and subsequently tested. Due to the diversity of

predictors, the measures and analysis procedures are discussed in detail. A presentation

and discussion of the results follows.

217

10.2 Descriptive statistics

The aim of this section is to examine the change that has occurred over time in the

entire group of supervisors / managers. First, the results of the paired sample t-tests

presented in Table 10.1 will provide an overall picture of the extent and direction of

change that has occurred over time in this group. Subsequently, Table 10.2 presents the

results of the independent sample t-tests between supervisors / managers and employees

at site 1. This permits a comparison of the change that has occurred within the

supervisory/ managerial group with that which has occurred within the group of

employees.

Table 10.1: Paired sample t-tests for the group of managers/supervisors at site 1

Variables

(N=51) Supervisors / managers

Time 1 Time 2 Change Scores

Mean (S.D) Mean (S.D) Mean (S.D)

General orientation to quality 6.01 (0.57) 6.17 (0.53) .16 (0.26)*** Improvement as part of the job 5.92 (0.84) 5.98 (0.84) .06 (0.19)** Intrinsic motivation 6.10 (0.64) 6.16 (0.50) .06 (0.57) Higher Order Need Strength 6.01 (0.84) 6.18 (0.52) .17 (0.90) Quality awareness 6.07 (0.57) 6.09 (0.57) .02 (0.70) Organizational commitment 5.32 (0.75) 5.58 (0.80) .26 (0.69)*** Theory X 4.02 (1.12) 3.84 (1.13) -.19 (0.98) Theory Y 5.08 (0.88) 5.39 (0.63) .31 (0.64)*** Trust in colleagues 5.46 (0.85) 5.49 (0.89) .03 (0.80) Immediate superior participative style 4.95 (1.19) 5.02 (1.26) .07 (1.37) Immediate superior commitment to quality 4.87 (1.23) 5.01 (1.21) .14 (1.13) Management commitment to quality 5.18 (1.06) 5.42 (0.88) .24 (0.95)*

* T-test difference in means between Time 1 and Time 2 significant at < than .1 level T-test difference in means between Time 1 and Time 2 significant at < than .05 level T-test difference in means between Time 1 and Time 2 significant at < than .01 level

Overall, the results' indicate that changes over time have been in the hypothesized

direction. There has been a significant positive change in general orientation to quality,

Paired sample t-tests were also conducted on 1st level supervisors / managers and the results were broadly similar.

218

improvement as part of the job, organizational commitment, subscription to a Theory Y

view of employees and to a lesser extent, perceived management commitment to

quality. Supervisory subscription to Theory X has shifted in a negative direction as

would be expected although this is not significant.

An interesting result is the lack of significant change in supervisors' / managers

perceptions of the behaviour of their immediate boss. In other words, there has not

been a significant positive shift in supervisors' perception of their superior's

participative style and commitment to quality. Given the cascading process, one would

expect a positive change in the way supervisors viewed the behaviour of their

immediate boss. Overall, this has not occurred. One explanation may be the

implementation process. As discussed in chapter 2, the cascading process was uneven

within the group of managers/ supervisors. Furthermore, some individuals resisted

cascading the intervention to their subordinates. Consequently, it is not surprising that

a significant positive change has not occurred within this group as a whole.

Overall, the results indicate that some significant change has occurred within this group

between time 1 and time 2. However, from these results, it is not possible to ascertain

whether this change is due directly to the intervention.

In view of the proposition that change at employee levels is dependent upon change

within the managerial hierarchy, the next step would be to compare the group of

supervisors/ managers with the group of employees. Table 10.2 presents the results.

219

Table 10.2: Independent t-tests between supervisors/managers and employees at site 1

Variables

(N=51) Supervisors / managers

Time 1

Mean (S.D)

Time 2

Mean (S.D)

Change scores

Mean (S.D)

General orientation to quality 6.01 (0.57)-H-+ 6.17 (0.53)-H-F .16 (0.26)*** Improvement as part of the job 5.92 (0.84)+-H- 5.98 (0.84)+++ .06 (0.19)**+ Intrinsic motivation 6.10 (0.64) 6.16 (0.50) .06 (0.57) Higher Order Need Strength 6.01 (0.84) 6.18 (0.52)-H-+ .18 (0.90) Quality awareness 6.07 (0.57) 6.09 (0.57) .02 (0.70) Organizational commitment 5.32 (0.75) 5.58 (0.80) .26 (0.69)*** Trust in colleagues 5.46 (0.84) 5.49 (0.89) .03 (0.80) Immediate superior participative style 4.95 (1.19) 5.02 (1.26) .07 (1.37) Immediate superior commitment to quality

4.87 (1.23) 5.01 (1.21) .14 (1.13)

Management commitment to quality 5.18 (1.06) 5.42 (0.88) .24 (0.95)* Improvement in commitment to quality 4.80 (0.88) Improvement in quality climate 5.26 (0.80)-H-+ Perceived benefit of intervention 5.09 (1.07)+++ Appropriateness of intervention 5.41 (0.90)+1-F Reinforcement of intervention 4.75 (1.10)

Variables

(N=165) Employees

Time 1

Mean (S.D)

Time 2

Mean (S.D)

Change scores

Mean (S.D)

General orientation to quality 5.62 (0.82)+++ 5.78 (0.75)+++ .16 (0.46)*** Improvement as part of the job 4.45 (1.18)+++ 4.58 (1.19)-FH- .13 (0.53)***+ Intrinsic motivation 6.20 (0.65) 6.24 (0.68) .04 (0.66) Higher Order Need Strength 5.87 (0.81) 5.88 (0.92)-H-+ .01 (0.88) Quality awareness 5.99 (0.67) 5.95 (0.65) -.04 (0.59) Organizational commitment 5.32 (0.98) 5.51 (1.00) .19 (0.74)*** Trust in colleagues 5.67 (0.93) 5.68 (0.91) .01 (0.76) Immediate superior participative style 4.98 (1.11) 5.15 (1.09) .17 (1.07)** Immediate superior commitment to quality

4.91 (1.25) 5.04 (1.19) .13 (1.12)

Management commitment to quality 5.10 (1.01) 5.27 (1.03) .17 (0.86)** Improvement in commitment to quality 4.84 (1.06) Improvement in quality climate 4.63 (1.22)+++ Perceived benefit of intervention 4.20 (1.31)+++ Appropriateness of intervention 4.99 (1.26)+++ Reinforcement of intervention 4.55 (1.27)

* T-test difference in means between Time 1 and Time 2 significant at < than .1 level ** T-test difference in means between Time 1 and Time 2 significant at < than .05 level *** T-test difference in means between Time 1 and Time 2 significant at < than .01 level

T-test difference in means between supervisors and employees significant at < than .1 level ▪ T-test difference in means between supervisors and employees significant at < than .05 level +++ T-test difference in means between supervisors and employees significant at < than .01 level

220

In comparing the two groups; supervisors and employees, several interesting findings

emerge. First, the significant changes that have occurred in the group of supervisors

have been mirrored in the group of employees. In both groups, there has been

significant positive changes in general orientation to quality, improvement as part of the

job, organizational commitment and perceived management commitment to quality. At

time 1, supervisors were not found to be significantly different from employees on a

range of measures (the two dimensions of commitment to improvement are the

exception). Furthermore, at time 2, there were no significant differences found between

supervisors and employees in their perception of the behaviour of their immediate boss

regarding participative style or commitment to quality. This may indicate an uneven

cascading effect where rather than cascading throughout the managerial hierarchy prior

to the lower levels, the cascading approach has occurred throughout certain parts of the

organization but not others. Further support for this is found in the absence of any

significant difference between employees and supervisors in their perception of their

immediate boss reinforcing the intervention. One could reasonably expect that if the

effects of the intervention cascaded throughout the managerial hierarchy in the initial

instance, then, managers/supervisors would be significantly more positive in their

perception of reinforcement of the intervention by their immediate boss. This has not

occurred.

In terms of the TQM intervention, a more optimistic result is the perceived differences

between the two groups in their assessment of the intervention, with supervisors being

more positive. Given the time span between the intervention and the collection of data,

the group of supervisors would be more proximal to the intervention and would have

greater experience of it, consequently, this result is not surprising.

221

To summarise, significant change has occurred between time 1 and time 2 in the group

of managers/ supervisor along some dimensions. From a TQM perspective, one could

argue that a significant positive change in Theory Y and a parallel (although not

significant) one in Theory X is a move in the right direction and possibly a requirement

for subsequent behavioural change. An interesting finding is that the significant

positive changes that occurred over time in the group of managers/supervisors are also

found in the group of employees. Previous research found that managerial attitudes and

behaviour after human relations training did not alter in the short term but significantly

changed after 18 months (Hand, Richards and Slocum, 1973). In this study, it may be

the case that supervisory attitudes and behaviour need to be stimulated or reinforced by

other mechanisms in the organization in order to take hold. Consequently, it cannot be

ruled out that in the longer term, supervisory attitudes and behaviour may alter

significantly along a wider range of dimensions.

Adopting a critical perspective, one could argue that little change has occurred between

time 1 and time 2. However, the interpretation of the extent of change that has occurred

is dependent upon how much change could be expected given the nature of the

intervention and the way it was implemented. This issue is pursued in the concluding

chapter.

222

10.3 An hypothesized model of supervisor participative style and commitment to quality

The hypothesized model of supervisor participative style and commitment to quality is

presented in Figure 10.1. The predictors of supervisor participative style and

commitment to quality can be grouped into the following two broad categories:

supervisor attitudes and supervisor perceptions of their immediate boss (perceived

superior reinforcement of the intervention, perceived superior participative style and

commitment to quality). The discussion begins with supervisory attitudes.

Supervisory attitudes This category contains four predictors: Theory X and Y

assumptions, perceived benefit of the intervention and behavioural commitment to the

intervention. Theory Y and X are hypothesized to have a positive and negative impact

respectively, on supervisors' participative style. While McGregor's (1960, 1967)

conceptualization of Theory X and Y managerial philosophies has exerted considerable

influence on managerial thinking, there is a paucity of empirical research linking these

views to managerial behaviour. McGregor (1960) believed that a manager's personal

beliefs about the nature of man exert significant influence over a manager's actual

behaviour when directing others. Managers accepting a Theory X view would be more

likely to deal with subordinates in an authoritative manner and are unlikely to behave in

a participative fashion. On the contrary, managers accepting a Theory Y view are more

likely to deal with subordinates in a participative manner reflecting the broad

assumption that employees can make an important contribution, are willing to accept

responsibility and are capable of exercising autonomy.

223

224

inte

rven

tion

Subo

rdin

ate

grou

p m

ean

resp

onse

Hyp

othe

siz e

d m

odel

of s

uper

viso

r pa

rtic

ipat

ive

styl

e an

d co

mm

itmen

t to

qual

ity

Perc

eived

imm

edia

te

Beh

avio

ural

com

mitm

ent

Perc

eive

d be

nefit

of t

he

inte

rven

tion

Fiman (1973) investigated the link between supervisory Theory X and Y assumptions

and their behaviour. He found a negative relationship between Theory X assumptions,

consideration and Theory Y behaviours, and a positive link between Theory X

assumptions and initiating structure behaviour. The relationship between Theory Y

assumptions and the measured behaviours was the exact opposite of that found for

Theory X assumptions. These relationships were found using subordinate perceptions

of supervisors' Theory X and Y assumptions 2 and subordinate perceptions of their

supervisors' behaviour. However, Fiman reports that these relationships were

altogether less clear when supervisors' own responses were used to represent Theory X

and Y assumptions and consideration, initiating structure and Theory Y behaviours.

McGregor (1967) stated that Theory X and Y were not opposite ends of the same

continuum but rather different cosmologies. Accordingly, one would not expect to find

significant positive or negative correlation between Theory X and Y. In this study, a

significant negative correlation of -.40 was found and this result is consistent with

previous findings (see for example, Jacoby and Terborg who report a negative

correlation of -.55). Thus, this may question the idea that Theory X and Y are distinct

and separate set of assumptions. Rather, it would seem that they are bipolar.

Supervisory behavioural commitment to the intervention is also hypothesized to have a

positive impact on supervisory participative style. In other words, the more committed

the supervisors are to the intervention, the more participative their style of managing.

The rationale for this is twofold. First, behavioural commitment to the intervention

taps supervisory perceptions of the extent to which they keep subordinates well

informed and involved in the intervention. Consequently, supervisors who involve

2 The piloting of the questionnaire in my research included subordinate perceptions of their supervisors' Theory X and Y assumptions. Respondents reported difficulty in answering these statements and they were subsequently omitted from the questionnaire

225

subordinates in the intervention are displaying a participative oriented behaviour which

should be perceived by subordinates as such. Second, one of the objectives of the

intervention is a shift toward a greater participative style and thus, supervisors who are

behaviourally practising the principles of the intervention and who view it as an

important part of their job are more likely to be perceived as participative. Overall,

supervisory reported behavioural commitment to the intervention is hypothesized to

have a positive impact on how employees perceive the participative behaviour of the

supervisor.

In addition, perceived benefit of the intervention is hypothesized to have a positive

impact on the participative style of the supervisor. In other words, the greater the

perceived benefit of the intervention, the greater the likelihood that supervisors will

shift toward a more participative style of managing. Hill (1995) argues that the

likelihood of a successful TQM change effort will increase if it is congruent with the

self interest of individual managers. Thus, if managers / supervisors perceive that they

will benefit from the intervention, they will be more likely to change their behaviour.

Supervisors' perception of the behaviour of their immediate superior. This

category contains three predictors. The first two relate to the supervisor's perception of

their immediate superior's participative style and commitment to quality. These two

predictors have been used in earlier models as antecedents of team orientation

(participative style) and commitment to improvement (superior commitment to quality).

In this model, these perceptions are of a supervisor's superior rather than employees

perceptions of their supervisor as used previously. It is hypothesized that a supervisor's

perception of the participative style of their immediate superior will have a positive

impact on the supervisor's own participative style. Similarly, a supervisor's perception

of their immediate superior's commitment to quality is hypothesized to have a positive

226

effect on the supervisor's own commitment to quality. The rationale for this link is

similar to that put forward in chapter 6 where it was hypothesized that a supervisor's

commitment to quality would have a positive impact on employee commitment to

improvement.

The final predictor in this category is a supervisor's perception of the degree to which

his/her immediate superior is reinforcing the intervention. It is hypothesized that the

more a supervisor perceives his/her immediate superior as reinforcing the intervention,

the more likely it is that the supervisor will be participative in his/her style of managing

and committed to quality.

10.4 Measures and analysis procedures

This section presents the new measures applicable solely to individuals in a supervisory

position which have not been discussed in previous chapters. In addition, as the

predictors outlined in the model come from two categories and are measured

differently, these are briefly discussed. Following from this, the analysis procedures

are presented.

Theory Y This four item scale was adapted from a 12 item scale developed by

Jacoby and Terborg (1975). The authors found the longer 12 item scale to have a high

internal consistency (r=.85) and a test-retest reliability of .68 based on data gathered

from supervisors and non supervisors. The four item scale used in this study taps

supervisors' agreement or disagreement on general statements about employees; the

degree to which individuals are self motivated and controlled; the extent to which

employee potential is greater than typically recognized by organizations; the degree to

227

which the average person finds work a source of satisfaction and; the extent to which

employees are capable of exercising autonomy and independence on the job. The alpha

coefficients for this measure using the overall sample of supervisors / managers was .71

at time 1 and .63 at time 2.

Theory X This four item scale was adapted from a 24 item scale developed by

Jacoby and Terborg (1975) which exhibited a high internal consistency of .77 and a

test-retest reliability of .59. The four items used in this study tapped supervisors'

agreement or disagreement on the following: the extent to which an average person

wishes to avoid responsibility; the degree to which most employees do not possess the

potential to be self starters on the job; whether giving greater independence to most

employees would be bad for the organization and; whether an increase in pay is enough

to overcome people's inherent dislike of work. The alpha coefficients for this scale

using the overall sample of supervisors / managers was .69 at time 1 and .74 at time 2.

Behavioural commitment to the intervention This five item measure taps two

dimensions of a supervisor's commitment to the TQM intervention. The first

dimension relates to their behaviour towards their subordinates in terms of keeping

them well informed and involving them in the intervention. The second dimension taps

respondents' perceptions of the degree to which they are practising the principles of the

intervention, the extent to which they perceive the intervention as an important part of

their job and whether they think that their colleagues would say that the respondent

himself/herself is committed to the intervention. This scale exhibited a high level of

reliability of .86 at time 2.

In addition to the new measures discussed above, the hypothesized model contains a

number of measures that have been discussed in previous chapters. For example,

228

perceived benefit of the intervention was discussed in chapter 5. In examining the

antecedents of team orientation (chapter 4), the participative style of the immediate

supervisor was presented. Perceived superior commitment to quality was discussed in

chapter 6.

The first stage in the analysis procedures involved the matching of subordinates to their

immediate supervisor. This was necessary to create the two dependent variables:

supervisory participative style and commitment to quality. These two variables

represent the mean subordinate group response on two dimensions of their immediate

supervisor's behaviour: participative style and commitment to quality. The advantage

of using the subordinate group mean response is twofold: first, it represents a more

objective measure than asking supervisors how participative they are in their style of

managing or how committed to quality they are. Second, as shown in the previous

chapters, it is employees' perception of the behaviour of their immediate boss that is

important in affecting employee attitudes and behaviour. The argument here is that

employee perceptions more accurately represent their reality. After all, an employee

can only react to what he/she perceives.

In investigating superior-subordinate relationships, there are a variety of levels of

analysis that may be adopted (see Yammarino and Dubinsky, 1992 for a review). Some

researchers assert that superior-subordinate relationships are on a one-to-one basis

where the superior displays a different style toward individual subordinates within the

work group (Dansereau, Graen and Haga, 1975). Thus, a superior may act

participatively toward one subordinate and behave more autocratically toward another.

An opposing view is asserted by other researchers (Schriesheim and Kerr, 1977) which

assume that a superior has a similar relationship with the subordinates in the work

group. The key debate rests on whether one assumes heterogeneity or homogeneity of

229

leader behaviours. Cummings (1975) argues that leaders do not behave differently

toward subordinates for two reasons: " (a) equality considerations aimed at countering

accusations of preferential treatment, and (b) time and energy costs associated with the

diagnosis necessary to behave heterogeneously" (p 184).

Schriesheim (1979) found a very strong relationship between how individuals perceived

the behaviour of the leader toward the work group and how they perceived the

behaviour of the leader toward them as individuals. He argues "that although leaders

act somewhat differently toward individual subordinates, each leader also has a more

general behavioral pattern that subordinates recognize and respond to" (p346). This

view is closely aligned to the one adopted here. The mean response of employee

perceptions of their supervisor's behaviour assumes that there is a basic degree of

consensus in how the group perceives the behaviour of their supervisor. Thus, a

supervisor may adopt a general participative style and within this behave more or less

participatively toward individual subordinates.

The analysis was restricted to first line supervisors; that is, the immediate level above

employees. Consequently, the sample size is quite small (n=27). However, many of

the previous studies investigating issues of supervisory / managerial behaviour have

utilized similar sample sizes ( see for example, Baum, Sorensen and Place, 1970;

Decker, 1982; Fiedler and Mahar, 1979; Hand and Slocum, 1972).

As the sample size is quite small the number of predictors in the model was kept to a

minimum. Consequently the only control variable included is the mean response of the

subordinate group as to the degree to which they perceive their supervisor as

reinforcing the intervention.

230

In terms of the analysis, the mean subordinate group response to supervisory

participative style and commitment to quality at time 2 were separately regressed on the

other variables in the model using OLS regression. 3 The analysis was conducted on the

time 2 cross sectional data controlling for the dependent variable at time 1. Using the

change data was not possible for two reasons. First, there was some movement of

employees and supervisors between time 1 and time 2 thereby making it difficult to

calculate reliable group mean responses across the two occasions of measurement.

Second, the group mean response may not pick up changes at the individual level. For

example, if a group mean response remains stable over time, this may be due to

changes (positive or negative) at the individual level which may counteract each other.

The mean group responses of perceived supervisor participative style ranged from 3.35

to 6.14 and 2.50 to 6.11 for perceived supervisor commitment to quality. The number

of employees comprising the groups ranged from 2 to 10 with the majority of groups

containing the mean response of 5-6 employees.

3 Subordinates were matched to their immediate supervisor at time 1 to provide a mean subordinate group response to supervisory participative style and commitment to quality. Between time 1 and time 2, there was some movement of employees and supervisors between different work areas.

231

10.5 Results

Table 10.3 presents the results of the predictors of supervisor participative style and

commitment to quality controlling for the dependent variables at time 1. 4

Table 10.3: Predictors of supervisor participative style and commitment to quality

Predictors Supervisor

participative stylet Supervisor

commitment to quality'

Theory X -.48— -.30" Theory Y .12 .18 Perceived benefit of the intervention .19 .26" Perceived superior reinforcement of the intervention .3 1 k .26" Behavioural commitment to the intervention -.73 — -.64— Perceived immediate superior participative style .07 .'''Y;, ,i,„''l," ' ,iz

Perceived immediate superior commitment to quality / 0;:c/a.„0:7, .1 6

Adjusted R2 .58 .77 N 27 27

+= p<.10 = p<.05 = p<.01

t subordinate group mean response

The results highlight several significant findings. First, the significant predictors of

supervisor participative style and commitment to quality are broadly similar with one

exception: perceived benefit of the intervention has a significant impact on perceived

supervisory commitment to quality but not on perceived supervisor participative style

(although the beta coefficient is high). Second, Theory Y was not found to have a

significant impact on participative style or commitment to quality. Also, a supervisor's

The regression analysis was also conducted without controlling for the dependent variable (mean subordinate group response on the two dimensions of supervisory behaviour) at time 1. The results yielded the same significant predictors as controlling for the dependent variables at time 1.

232

perception of their own superior's participative style (or commitment to quality) was

not found to significantly affect the supervisor's own participative style (or

commitment to quality). Finally, contrary to the hypothesis, supervisory behavioural

commitment to the intervention has a significant negative impact on his/her perceived

participative style. In other words, the greater their commitment to the intervention, the

less they are perceived by their subordinates to be operating along participative lines or

to be committed to quality.

10.6 Discussion

In terms of the model, the predictors, with one exception, have an impact in the

hypothesized direction although this is not always significant. Overall, the predictors

of supervisor participative style and commitment to quality are very similar.

Theory X

The negative significant impact of Theory X assumptions on participative style is

entirely consistent with McGregor's (1967) thinking. The result indicates that the more

a supervisor accepts Theory X assumptions, the less likely he/she will be perceived by

subordinates to behave in a participative manner This lends support to McGregor's

(1967) proposition that managerial assumptions about human nature have a significant

impact on the way individual managers interact with their subordinates. As the analysis

is cross sectional in nature, what it shows is that participative supervisors tend to accept

Theory X assumptions less than non participative or less participative supervisors.

However, the results do not show that changes in Theory X assumptions will lead to

changes in participative style.

233

One of the methodological weaknesses of Fiman's (1973) study is the use of

subordinate perceptions of his/her supervisor's Theory X and Y assumptions and

dimensions of his/her supervisor's behaviour. In using the subordinate as rater of both

assumptions and behaviour, this method may be open to bias thus attaining greater

consistency between perceptions of assumptions and behaviour. This study overcomes

this problem by utilizing supervisors' Theory X and Y assumptions and employee

perceptions of supervisory behaviour. Consequently, this is a more direct and stringent

test of McGregor's (1960) attitude-behaviour linkage which hypothesizes that the

supervisor's attitude (not the employee perception of that attitude) would affect his/her

behaviour.

Theory Y

In contrast to the significant negative effect of Theory X on supervisory participative

style, Theory Y was found not to have a significant positive impact on participative

style. This is contrary to the findings of Fiman (1973) who found a positive

relationship between Theory Y assumptions and consideration and Theory Y

behaviours. Moreso, the absence of a significant impact in this study goes against the

implicit assumption underlying much of the HRM literature which emphasizes a

Theory Y view of employees presumably resulting in Theory Y type behaviours. In

terms of competing predictors, the results suggest that Theory X is a more significant,

albeit negative, predictor of participative style than Theory Y. It is worth considering

several possible explanations for this result.

The first possible explanation concerns the measurement of Theory Y assumptions; that

is, the selection of the four items from Jacoby and Terborg's original 12 item scale.

The argument would be that the four items selected may not be expected to relate

directly to participative behaviour; that is, the four items do not specifically tap

234

participative type assumptions. 5 This explanation is unlikely as the same argument

could be levelled against the items measuring Theory X assumptions. The second

possible explanation has to do with a more rigorous methodology in terms of using a

supervisor's attitude and employee perception of their behaviour. While this seems a

plausible rationale, it is unlikely given the significant impact of Theory X.

A third possibility is that beyond a certain point, greater subscription to Theory Y has

no further effect on Theory Y type behaviours. While potential explanations have been

raised, it is not possible to speculate any further given the paucity of research

examining the link between Theory X and Y and behaviours.

Superior reinforcement of the intervention

A supervisor's perception of their immediate superior reinforcing the intervention has a

significant effect on the supervisor's commitment to quality and to a lesser degree on

participative style. This is consistent with the recommendations from leadership

(House, 1968) and human relations (Hand and Slocum, 1970) training. The issue here

is the transferability of the training back to the workplace. Hand and Slocum (1970)

argue that the superior must act as a reinforcer so that the principles learned on the

training program become integrated into the operating procedures of the organization.

Hand et al. (1973) in their study of human relations training found no significant

change in managerial attitudes and behaviour after 3 months. However, significant

change was evidenced after 18 months. The authors suggest that for a training program

to take effect, the organization's decisions need to reinforce the attitudes learned on the

training program on an ongoing basis. This is consistent with prior research

(Fleishman, 1955; House, 1968) which showed the ineffectiveness of training in

inducing attitudinal and behavioural change when such changes are not supported in the

5 For example, the average person prefers to have greater influence in decision making than typically given in organizations or most employees would welcome the opportunity for greater participation

235

work environment. With particular reference to human relations training, Hand and

Slocum (1972) argue "it is difficult to produce a change in individual behaviour

toward being more considerate to the needs of others if the culture is imbedded with

other values" (p 416).

The importance of reinforcement has implications for TQM change efforts. This study

has highlighted the importance of one source of reinforcement, namely, the supervisor's

superior. Most of the writings on TQM emphasize leadership, education and training as

the primary mechanisms to affect change (Deming, 1986; Oakland, 1989). However,

Hill (1995) argues that it may be necessary to use additional mechanisms such as

reward and punishments in order to reinforce the desired change. In the present

organization, management, assumed that TQM training, appropriate resources and

arrangements would lead to the desired culture change. By the end of the second round

data collection, management realised that additional reinforcements would be necessary

to ensure attitudinal and behavioural change. Consequently, while the superior may be

an important source of reinforcement, it may not be sufficient on its own.

Perceived benefit of the intervention

Perceived benefit of the intervention has a significant positive impact on supervisory

commitment to quality Thus, the greater the perceived benefit, the more a supervisor

is perceived to be committed to quality. This has practical implications for those at the

top of the organization in their efforts to elicit quality oriented attitudes and behaviours

from supervisors and managers. There is a need to ensure that individual managers are

aware and informed of the future benefits to be gained from implementing TQM. The

perceived benefits may take the form of greater involvement and influence in decision

making both upwards and horizontally or it may take the form of career advancement

through gaining attention of senior managers as a result of performing well in terms of

236

TQM (Hill, 1995). The importance of perceived benefit has already been highlighted in

terms of having a significant impact on improvement as part of the job and to a lesser

degree on intrinsic motivation at employee level. Thus, perceived benefit has a role to

play in affecting change at all levels in the organization.

Behavioural commitment to the intervention

The impact of supervisory commitment to the intervention on the two dependent

variables is contrary to the hypothesized relationship. One would have expected greater

commitment to the intervention to have a positive impact on perceptions of supervisory

behaviour. The result appears to be counter intuitive; the more committed supervisors

are to the intervention the less supervisors are perceived to be participative and

committed to quality by their subordinates. The initial explanation may be a

measurement issue; that is, there is a flaw in the measurement of behavioural

commitment to the intervention.

One way to test this is to investigate other attitudes measured at the supervisory level.

For example, in predicting employee perceptions of their supervisors' commitment to

quality, one could include the supervisor's own reported general orientation to quality

as a predictor. One would expect that the stronger a supervisor's (self reported) general

orientation to quality, the more his/her subordinates would perceive the supervisor to be

committed to quality. If this was found to be the case, it would suggest that there is a

flaw in the measurement of supervisor behavioural commitment to the intervention.

The results 6 indicate that the stronger a supervisor's general orientation to quality, the

less they are perceived by their group of employees to be committed to quality (beta

coefficient -.35, p<.10). What this result suggests is that the measurement of

6 This was tested by replacing a supervisor's self reported behavioural commitment to the intervention with their self reported general orientation to quality as a predictor of perceived (employee group mean response) supervisor commitment to quality.

237

commitment to the intervention is not flawed. Rather, a more likely explanation points

to a fundamental difference between a supervisor's perception and evaluation of his/her

own behaviour and that of how his/her subordinates perceive and evaluate the

behaviour of the supervisor.

This result highlights the problem of using different types of measurement. Rosen

(1969) obtained a similar result in linking supervisory attitudes to work group

productivity. He found that supervisor's self report characteristics were not related to

work group productivity whereas subordinate's perception of the supervisor were

related to productivity. Thus, a potential explanation for the result found in this study

may be a result of differing perceptions; a supervisor perceives his/her own behaviour

differently to how employees see it. This raises a further issue of why this occurs for

supervisors' self report measure of commitment to the intervention and not with the

other self report supervisory measures used. The distinguishing difference between

supervisory commitment to the intervention and the remaining supervisory self report

measures is that commitment to the intervention is directly tapping the individual

supervisor's behaviour. The remaining measures tap: general views of employees

(Theory X and Y), perception of the intervention (perceived benefit) and the

supervisor's perception of his/her own superior's behaviour (participative style and

commitment to quality)

Pursuing the explanation that supervisors perceive their own behaviour differently

(more favourably) to how their subordinates see it, why might this be the case? Why is

it that the more supervisors report being committed to the intervention, the less they are

perceived to be participative and committed to quality by their group of subordinates?

This may be the result of the interaction of two processes. First, subordinates may be

inclined to adopt a more critical view of their supervisor's behaviour (especially in the

238

dimensions that relate to involving subordinates) than the view held by the supervisor.

Second, supervisors have a central role to play in the intervention, consequently they

may be more likely to inflate their commitment to the intervention as they have more at

stake.

Superior behaviour

A supervisor's perception of their superior's behaviour (either in terms of participative

style or commitment to quality) does not have a significant impact on the supervisor's

own behaviour. This goes contrary to the findings of a previous chapter where

employee perception of their supervisor's commitment to quality had a significant

impact on the employee's own commitment to quality. Thus, it would seem that the

role expectations of a superior is only operative for employees. This is similar to the

finding of Scott and Bruce (1994) where role expectations had a significant effect on

the behaviour of technicians but not on engineers or scientists. The authors offer a

potential explanation in that high education, high independence and status equality of

the engineers and scientists with managers reduce their receptivity to leader role

expectations. A similar explanation may operate in this study.

However, the explanation may also lie with the use of different forms of measurement.

To test out the explanation offered by Scott and Bruce (1994), one would need to use

supervisor's self reported behaviour. In this case, it is not the supervisors' self reported

participative style and commitment to quality that is used but rather employee

perceptions of this behaviour. As already mentioned, employees and their supervisor

may have different perceptions of the supervisor's behaviour and this may account for

the absence of a significant relationship.

239

In summary, for organizations implementing TQM, it is important for those in

supervisory positions to perceive the intervention in beneficial terms and also to

perceive their superior as reinforcing the intervention. These perceptions do have an

impact on how supervisors' own behaviour is in turn perceived by their subordinates.

10.7 Conclusions

This chapter set out to examine what change had occurred within the

supervisory/managerial group. Subsequent to this, two dimensions of the behaviour of

first line supervisors; participative style and commitment to quality, were investigated.

These dimensions of behaviour were measured by using the mean subordinate group

response. This method could be considered to be more objective than asking

supervisors about their own behaviour and in addition, as previous chapters have

shown, it is pertinent in that it represents employee reality or perceptions of it.

The results suggest that Theory X is a better predictor of supervisory behaviour than

Theory Y. The importance of superior reinforcement of the intervention in affecting

supervisory behaviour was highlighted. Supervisors' perception of their superior's

behaviour was not found to have any significant effect on the supervisor's own

behaviour. Contrary to the hypothesis, a supervisor's self reported commitment to the

intervention had a significant negative effect on their behaviour as perceived by their

subordinates.

The subsequent chapter integrates the findings of this and previous chapters by making

an overall assessment of the intervention. It does so by examining the intervention

using five different criteria. The first criterion has been the primary focus of this thesis,

namely, the impact of the intervention on key elements of TQM. This is supplemented

240

with an examination of the impact of the intervention on perceived performance

improvement and employee involvement outcomes. In addition, the intervention is

evaluated on the extent to which it achieved its goals. Finally, the impact of the

intervention is examined from a TQM perspective; how would TQM proponents view

the effect of the intervention?

241

Chapter 11: An Overall Assessment of the TQM Intervention

11.1 Introduction

Three distinct strands of inquiry were pursued in the thesis. The initial focus was on

investigating the impact of the TQM intervention on some of the core elements of

TQM; team orientation, intrinsic motivation, general orientation to quality and

improvement as part of the job. The second strand examined the predictors of

employee participation in and assessment of the TQM intervention. These two lines of

inquiry focused exclusively on employees. In view of the importance of supervisory

behaviour, the final line of investigation concentrated on supervisors / managers.

However, one question remains, how does one assess the overall impact of the

intervention? This chapter addresses this key question by examining the impact of the

intervention using a number of different criteria.

Prior to pursuing the issue of assessment, it is worthwhile considering some of the key

findings of this study. This discussion excludes the impact of the intervention on the

core elements of TQM which is reviewed later as one of the criteria by which the

intervention may be assessed.

Who participates in the TQM intervention? This question was addressed in chapter 8

and involved investigating the predictive power of a priori employee attitudes and

perceptions on their subsequent participation in the intervention. The single most

important determinant of employee participation in the intervention was the behaviour

of the supervisor. In other words the participative style of the supervisor at time 1

(prior to the intervention) was found to be a significant predictor of the extent to which

their subordinates participated in the intervention. It would seem that supervisors who

242

are generally more participatively orientated are more willing to cascade the

intervention to employees as it is consistent with their noimal style of managing.

Unlike other employees participation or involvement programs where the mechanisms

by which employees may volunteer to participate are independent of the supervisor, the

cascading of a TQM intervention relies on the behaviour of the supervisor to affect

employee participation. This, in theory, overcomes the obstacle confronting QCs by

integrating supervisors into the improvement process. In doing so, it is potentially

damaging to employee participation if a supervisor's behaviour does not facilitate such

participation.

Chapter 9 followed a similar line of inquiry and investigated what factors, if any,

predispose employees to assessing the intervention more or less positively. One

significant finding to emerge from this chapter was the difference in the predictors of

participation in and assessment of the intervention. Thus, while employee behaviour

(participation) is dependent upon the behaviour of the supervisor, employee attitudes

(toward the intervention) are not only influenced by supervisory behaviour but also by

other attitudes held by employees prior to the intervention. In addition the results

suggest that employees do carry with them their previous experience of a participative

program which influences their assessment of future participative programs.

These results directly challenge the validity of the underlying assumption in TQM that

employees will react similarly to TQM. Moreso, it questions the taken for granted view

"that employees will welcome, be committed to, and benefit from... [TQM]" (Wilkinson

and Willmott, 1995, p12).

243

Chapter 10 concentrated on supervisory behaviour in view of its importance in affecting

employee attitudes and behaviour. For the entire group of supervisors / managers, there

was a lack of significant positive change in their perceptions of the behaviour

(participative style and commitment to quality) of their immediate superior. This

suggests that within the managerial hierarchy, the cascading process was uneven; that

is, it occurred in some areas but not others. In terms of predicting the behaviour

(participative style and commitment to quality) of first line supervisors, perceived

benefit of the intervention and superior reinforcement of the intervention were found to

have a significant positive effect s at the individual level. Consequently, the supervisor's

superior is an important source of reinforcement. The significant effect of perceived

benefit suggests that organizations need to ensure that individual supervisors are aware

and informed of the benefits from implementing TQM.

Overall, the findings raise a broad issue that warrants discussion. The issue stems from

an assumption made in the earlier chapters that the TQM intervention variables are

exogenous in relation to the other variables in the model 2 . Subsequent findings of the

predictors of employee participation in and assessment of the intervention question the

validity of the assumption made. For example, in the team orientation model,

supervisory participative style was treated as an outcome of the intervention. However,

in chapter 8 and 9 respectively, supervisor participative style was found to be a

significant predictor of different aspects of the intervention; namely, employee

participation and perceived appropriateness of the intervention. A similar case is found

for organizational commitment. In chapter 7, it was treated as an outcome of the

intervention and in chapter 9, it was found to predict how individuals assessed the

intervention.

This applies more to supervisory commitment to quality than supervisory participative style. 2 This assumption applies to chapter 5 and 7.

244

While longitudinal research facilitates the attribution of cause and effect, these findings

highlight the difficulty in dealing with the issue of endogeneity particularly in

evaluation studies. An argument was put forward that, for example, organizational

commitment affected the assessment of the intervention and was subsequently affected

by the intervention. Thus, it was suggested that organizational commitment was both

an antecedent and a consequence of the intervention. Similarly, it was also found that

supervisor participative style was an antecedent and an outcome3 of the intervention.

What seems to emerge from this study is that care needs to be exercised in attributing

cause and effect. Organizational interventions do not occur in a vacuum, prior

experience, attitudes and behaviour of organizational members have an effect on the

intervention in question. Figure 11.1 presents the role of organizational commitment

and supervisor participative style in the process of change.

Figure 11.1: The role of organizational commitment and supervisor participative style in the process of change

Time 1 Time 2

Supervisory participative style

Supervisor participative style


Intervention


3 The evidence for treating supervisory participative style as an outcome is not as convincing as for organizational commitment. Nonetheless, for non participants in the intervention, their perception of . supervisory participatory style shifted in a positive direction between time 1 and time 2 (significant at 10% level).

245

To conclude that the process of change is accurately presented to the right of the dotted

line in Figure 11.1 does not do justice to the complexity of organizational change. In

other words, to argue that the intervention (cause) has an impact on organizational

commitment (effect) is not incorrect but only provides one half of the picture. A more

accurate representation would take into account (left of the dotted line) the impact of

organizational commitment (cause) on the intervention itself (effect). Therefore, it is

argued that Figure 11.1 in its entirety is more accurate in reflecting the complexities

involved in evaluating organizational change interventions.

Prior to discussing the criteria of assessment, the interpretation of the term impact

requires clarification. In particular, the levels at which it is used need to be

distinguished. It may be used to describe the effect of the intervention at the individual

level and also at the broader organizational or system level to describe the extent of

change that has occurred. This difference may be translated into two separate

questions. First, has the intervention had an impact on individual attitudes? And

second, has there been an overall improvement in the system? For example, the

intervention may have a significant effect on team orientation at the individual level but

there may not be an overall improvement in team orientation throughout the

organization as a whole. What this suggests is that the intervention in principle can

have an effect on team orientation and raises the possibility that greater time is needed

for the effects at the individual level to develop into an improvement at the system

level. Another scenario may be that the intervention has no effect at the individual

level but there has been a significant improvement at the system level. This would

seem to indicate that the positive change is not linked to the intervention in a direct

sense. In other words, this improvement may be an unintended consequence of the

intervention or due to other factors not related to the intervention.

246

Making a distinction between the impact of the intervention at the individual and

system level allows for a more finely tuned assessment of the intervention. Table 11.1

presents a summary of the impact of the intervention at the individual level and also

indicates whether there has been a significant positive change over time in the overall

sample of employees.

Table 11.1 shows that at the individual level, the intervention has a significant total

effect on all the variables. Consequently, the intervention can make a difference to

individual attitudes. In terms of answering the second question: has the intervention

been effective in leading to an overall improvement in the system, the results are not so

clear cut. Not all individual level effects have resulted in system improvements. For

example, while the intervention has a significant effect (at the individual level) on team

orientation and intrinsic motivation, there has not been an overall improvement. On the

other hand, the intervention has a significant effect on organizational commitment and

perceptions of supervisor participative style (at the individual level) and a significant

positive shift has also occurred over time.

247

Tab

le 1

1.1:

Sum

mar

y ta

ble

of th

e im

pact

of t

he T

QM

inte

rven

tion

to q

ualit

y

Supe

rvis

or

p arti

cipa

t ive

style

Org

aniz

atio

nal

com

mitm

ent

'->

Perf

orm

ance

im

prov

emen

t In

trin

sic

mot

ivat

ion

a-

Impr

ovem

ent

as p

art o

f the

jo

b

Gen

eral

or

ient

atio

n to

qua

lity

Team

or

ient

atio

n

1.

Parti

cipa

tion

in T

QM

in

terv

entio

n

2.

Ass

essm

ent'

of in

terv

entio

n

3.

Supe

rvis

ory

rein

forc

emen

t of

inte

rven

tion

0 z

0 z

0 z

248

11.2 Criteria of assessment

What criteria should be adopted in assessing the overall impact of the intervention?

Would it be sufficient to assess the intervention in terms of its impact on some of the

core elements of TQM? Campbell et al. (1970) argue that program effectiveness cannot

be appropriately measured in terms of a unitary or global variable. Instead, evaluation

studies need to employ multiple criteria. Each criterion has its own strengths and

weaknesses. For example, it would seem logical to assess the intervention in terms of

its goals; what it set out to accomplish. However, the intervention may not only have

multiple goals that are ambiguously stated but also, they may not be shared by those

responsible for the intervention. To compound the difficulty in using goals as a criteria

of assessment, the goals themselves or the way they are perceived may change during

the intervention (Golembiewski, Billingsley and Yeager, 1976).

Given the limitations of relying upon a single criterion to assess the impact of the

intervention, a multiple criteria approach is adopted. This type of approach is deemed

appropriate in view of the all encompassing nature of TQM. Furthermore, a multiple

criteria approach provides a more extensive method of ascertaining the overall impact

of the intervention. This is particularly important in view of the lack of evaluation

studies investigating the impact of TQM interventions.

Five criteria are used to assess the impact of the intervention. The first criterion

involves a re-examination of the intervention in terms of its impact both at an

individual and system level on some of the core elements of TQM. The second

criterion builds on the previous one by examining the effect of the intervention and the

core elements on perceived performance improvement. Subsequently, the impact of the

intervention is examined by looking at its effect on employee involvement outcomes.

249

Consistent with evaluative research, the extent to which the intervention achieved its

goals is the basis of the fourth criterion. Finally, the intervention is assessed from a

TQM perspective; how would TQM proponents or advocates view the effect of the

intervention?

11.2.1 Elements of TQM

The initial criterion selected to assess the intervention is its impact on some of the core

elements of TQM. Essentially, this involves a reexamination of the results presented in

chapters 5 (team orientation) and 7 (intrinsic motivation, general orientation to quality

and improvement as part of the job).

At the individual level, the impact of the intervention on team orientation was indirect;

supervisory reinforcement of the intervention had a positive effect on trust in colleagues

which in turn had a positive effect on team orientation. Thus, a link was found between

the intervention and the outcome of team orientation. Adopting a system view of

impact, was there an overall improvement between time 1 and time 2 in team

orientation?

In terms of a significant positive improvement in team orientation for the entire group

of employees, this has not occurred. In fact team orientation shifted in a negative

direction between time 1 and time 2. Thus, although the intervention had a significant

effect at the individual level, it was not effective in achieving a positive improvement in

the overall group of employees.

However, in view of the short time span of this study, one could plausibly argue that

assessing the impact of the intervention in terms of its positive effect on the entire

250

group of employees (regardless of their participation) is not only unfair but does not

reflect the process of change. Particularly in the early stages of the change process (as

in this study), there will be individuals who are participating in the intervention and

those who are not. Consequently, it may be more logical to examine the impact of the

intervention at a subsystem level; that is, the participants in the intervention. Pursuing

this line of reasoning, the evidence suggests that for participants (as a group) team

orientation has remained unchanged between time 1 and time 2.

In evaluating change interventions, it is worth distinguishing between intended and

unintended consequences. The negative shift in team orientation which can be

pinpointed to the group of non participants (5.51 at time 1 and 5.28 at time 2) may be

considered as an unintended consequence of the intervention. It was previously

suggested that this may be a result of non participants feeling excluded from the

intervention. This, of course, may be a temporal effect which will disappear once the

intervention has diffused throughout the organization.

In contrast to the indirect effect of the intervention on team orientation, the intervention

was found to have a significant direct effect on improvement as part of the job (table

7.3). Therefore, at the individual level, it would seem that the intervention can make a

difference. With regard to general orientation to quality, the intervention was found to

have a significant total effect (the majority of this effect being indirect) at the individual

level.

Has there been an overall significant improvement in general orientation to quality and

improvement as part of the job? From Table 11.1, the results indicate that there has

been a significant positive shift in these two elements. However, this may be largely

due to the method by which they were measured at time 1.

251

At the individual level, the intervention had a significant total impact on intrinsic

motivation. However, at the system level, there has not been a significant improvement

in intrinsic motivation. Adopting the same logic as outlined for team orientation, it

may be more appropriate to examine the impact of the intervention at the subsystem

level; participants in the intervention. For participants, intrinsic motivation has

remained unchanged between time 1 and time 2 (6.30 at time 1 and 6.36 at time 2). A

similar result is found for non participants (6.09 at time 1 and 6.11 at time 2).

Therefore, the intervention has not been effective in improving intrinsic motivation for

participants. As such, this result is similar to team orientation which remained

unchanged for participants between time 1 and time 2.

Using the impact of the intervention on some of the key elements of TQM, what kind of

assessment can be drawn from this? It is apparent that different assessments will be

contingent upon the level of impact. For example, at the individual level, the

intervention has a significant impact on the four elements of TQM. On the other hand,

(putting aside the two dimensions of commitment to improvement due to the method by

which they were measured at time 1) at a system level, the intervention has not had an

impact on any of the elements. Furthermore, taking a subsystem level (participants),

the intervention has not significantly improved team orientation or intrinsic motivation.

However, in making an assessment, one must consider the time span allowed for the

effects of the intervention to occur. Consequently, the counterargument to the

assessment at the system level would be that more time is needed for the individual

level effects to appear as an overall improvement in the system. However, even in the

short term, there is no indication that these individual level effects are beginning to

appear as an overall improvement in the subsystem; the group of participants. The

issue of time lags for effects to occur in pursued in the concluding chapter.

252

11.2.2 Performance outcomes

The primary motivator for organizations to adopt and implement TQM is the

hypothesized positive impact on organizational performance. While the espoused goal

of TQM is often stated as quality, in reality, the real objective is often to increase the

efficiency of the organization (Spencer, 1994). This is accomplished by reducing

organizational costs in terms of scrap, rework and inspection (Crosby, 1979). This is

further exemplified in more recent writings (Gevritz, 1991; Jacob, 1993) which

emphasize TQM's ability to induce cost savings, reduce personnel and increase profits.

In parallel, TQM has been cited as the mechanism for ensuring organizational

effectiveness (survival in the long run) in ever increasing competitive environments

(Eisman, 1992; Kiess-Moser, 1990). Grant at al. (1994) view TQM as encompassing

the long term interests of employees, shareholders and customers. Improving quality

secures the future for the organization and its employees. It may also lead to lower

costs and greater customer satisfaction which in turn enhances the competitive position

of the organization which benefits its owners and employees. This view of the benefits

of TQM is very similar to that espoused by Deming4 (1986).

In spite of the widely presumed benefits of TQM, there is little (if any) systematic

research which investigates the effects of TQM in terms of hard criteria measures of

performance (see Mohrman et al. (1995) for the impact of TQM on financial

performance criteria). Anecdotal evidence exists of successful TQM companies (see

Grant et al., 1994 and; Blackburn and Rosen, 1993) and Hill (1991b, 1995) presents

case study evidence of the benefits of TQM. While Hill (1995) acknowledges that the

4 Deming argues that quality improvement will lead to cost reduction which in addition to improved quality will meet the needs of the customers. This will lead to greater market share which ultimately will lead to lower prices and improved competitiveness.

253

organizations in his study may be exceptional in terms of their willingness and

presumably their capability of committing resources to the implementation of TQM, the

other companies heralded as successful examples of TQM have been successful prior to

the introduction of TQM. Mohrman et al. (1995) in their survey of 1,000 Fortune

organizations investigated the impact of TQM on perceived outcomes; direct

performance outcomes of work processes; company profitability and competitiveness

and; employee satisfaction and quality of work life. They found a significant

correlation between core TQM practices5 and all three outcomes. For manufacturing

organizations, a significant relationship was found between core TQM practices and

market share. The authors acknowledge the limitations of their study, in particular its

cross sectional nature and its reliance on the perceptions of a single senior person in

each organization. Consequently, from the evidence available on the impact of TQM

on organizational performance, it is not possible to discern whether TQM leads to

improved organizational performance or whether successful companies (to begin with)

have the slack resources available to invest and sustain a successful TQM effort.

The lack of empirical research investigating the effects of TQM on hard performance

criteria is not only indicative of TQM research to date but applies equally to the broader

context of organizational change interventions. Porras and Berg (1978) in their search

covering 160 change interventions found only 20 evaluative research studies that

examined organizational and work group change in terms of hard criteria. The issue of

hard criteria measures has again come to the fore in terms of the impact of Human

Resource Management (HRM). Guest (1992b) argues that research efforts to link

HRM and organizational outcomes have been hindered by inherent research problems

and thus progress has been limited.

5 These practices include quality improvement teams, quality councils, cross functional planning, process re-engineering, work simplification, customer satisfaction monitoring and direct employee exposure to customers.

254

The difficulty in linking TQM to organizational performance is further compounded by

the fuzziness surrounding what exactly it is and its presumed relationship with HRM.

While Guest (1992b) argues that the need for TQM to be integrated with HRM is

increasingly being recognized, Blackburn and Rosen (1993) from their research on the

Baldrige Award winning companies state that HRM policies "were mutually

interdependent, congruent, and directed at supporting a total quality management

perspective throughout the corporation " (p.50). Putting aside the extraneous factors

which may affect performance, should any performance be linked to TQM, is this due

to the "hard" aspects of TQM or the "soft" HRM factors or a combination of both?

While some of the problems linking TQM to organizational performance at the

organizational level have been outlined, a different set of obstacles confront any

attempt to link TQM to conventional measures of work group and individual

performance. Factors outside an individual's control must be included as potentially

influencing an individual's performance. Waldman (1994) argues that despite the lack

of theoretical attention given to opportunity factors, there exists empirical research that

highlights the effects of factors outside the individual's control on his/her performance.

Consequently, in view of the inherent limitations of hard performance measures6 and

the difficulty in obtaining valid measures, perceptual measures of performance

improvement were collected.

Specifically, individuals were asked whether their own performance and that of their

work group had improved compared to the year before (prior to the intervention taking

place). In addition, they were asked whether improvements had been made in how they

6 While some hard (e.g scrap, rework) measures were available, they were not comparable across production groups. There was no hard performance measure available that was common to all groups of individuals.

255

did their job. While this method is far from ideal, it does attempt to go further than

linking the intervention to TQM outcomes as discussed in the previous section.

However, one could level the criticism of social desirability against this method in that

individuals would be more inclined to inflate their performance improvement and that

of their work group. As the majority of questionnaires were completed on a one to one

basis, efforts were made to reduce an individual's inclination to inflate perceived

performance improvement.

Therefore, the next criteria adopted to assess the impact of the intervention looks at the

effect of the intervention on (perceived) performance improvement. 7 The measure of

performance improvement tapped the extent to which individuals' perceive an

improvement in their job performance and that of their work area as well as the degree

to which improvements had been made in how they did their job.

Analysis procedures

Up to now, team orientation, intrinsic motivation, general orientation to quality and

improvement as part of the job have been treated as key legitimate outcomes of TQM.

However, one could argue that these are intermediate outcomes which subsequently

affect performance Therefore, the following analysis addresses two questions. First,

is there a link between the intermediate variables (team orientation, intrinsic

motivation, general orientation to quality and improvement as part of the job) and

perceptions of performance improvement? And second does the TQM intervention have

an effect on performance improvement? To do this, performance improvement, the

dependent variable, was regressed on the intermediate variables, the intervention

The items measuring performance improvement were factor analysed with the items measuring an improvement in quality climate and commitment to quality. The results presented in appendix 16 indicate the factorial independence of performance improvement from the other two measures.

256

variables (participation in and assessment of) and the usual set of demographic

variables. This analysis was conducted at time 2 on the employee 8 sample at site 1.

Results

Table 11.2 presents the results of the predictors of perceived performance improvement.

Table 11.2: Predictors of perceived performance improvement

Predictor variables

Time 2

Perceived Performance Improvement

Job Tenure -.04 Age -.01 Gender -.08 Length of service .01 Job Title 2 .13 Job Title 3 .06 Change in supervisor -.06 Change in job content .08 Change of jobs .16** Team orientation .12 Intrinsic motivation -.08 General orientation to quality .15+ Improvement as part of the job .00 Participation in the TQM intervention .19** Perceived benefit of the intervention .01 Perceived appropriateness of the intervention .05 Perceived supervisory reinforcement of the intervention .26***

Adjusted R2 .35 N 165

+ = p<.1 ** = p<.05 *** = p<.01

t Measured at time 2

8 The analysis was restricted to the employee sample so that team orientation and participation in the intervention could be included as predictors of performance improvement. These two predictors (as mentioned in earlier chapters) were not measured at supervisory level.

257

The results suggest that the impact of the intermediate variables on performance

improvement is at best exceedingly weak. In contrast, an individual's participation in

the intervention and their perception of supervisory reinforcement of the intervention

have a significant effect on perceptions of performance improvement. In other words,

an individual's behaviour and that of his/her supervisor significantly affect performance

improvement. What is interesting is that individual attitudes toward their work group,

job, quality and improvement do not have a significant effect on performance

improvement.

These results provide some support for Guest's (1992a) argument (in the context of

HRM) for the need to shift the emphasis placed on changing attitudes to include efforts

that are targeted at changing behaviour to improve performance It would seem from

the results here that behaviour is a more significant predictor of performance

improvement than attitudes. This is interesting in light of the findings of the evaluative

element of the thesis whereby participation in the intervention was not a significant

predictor of team orientation, intrinsic motivation, general orientation to quality and

improvement as part of the job. Rather, individuals' assessment of the intervention was

found to be a better predictor of these outcomes. With regard to perceptions of

performance improvement, the opposite holds true; participation is a more significant

predictor of performance improvement than assessment of the intervention.

Overall, the results highlight the importance of changing behaviour (in this case getting

individuals involved in the intervention) as the route to performance improvement.

This does not undermine the importance of attitudinal change but in the short term,

behavioural change may be more significant. Therefore, the practical implications

would be a dual approach to change with education and training serving as the primary

mechanisms to attitude change with changes to work design and participation in

258

continuous improvement as the way to induce behaviour change. Guest et al. (1993)

highlight the possibility that an indirect approach to behaviour change (via first

changing attitudes) may be less successful than a direct approach to behavioural

change. The results here highlight the importance of behaviour over attitudes in

affecting perceived performance improvement.

Thus, it would seem that the behavioural components of the intervention (employee

participation in and supervisory reinforcement of the intervention) have a significant

effect on perceived performance improvement. However, the results and their

implications must be interpreted cautiously due to the measurement of performance

improvement. What is required is the use of hard criteria of performance in order to

test more rigorously the links between behaviour, attitude and performance in the

context of TQM.

11.2.3 Other outcomes

The thesis particularly focused on some of the core elements of TQM. Again, in

assessing the impact of the intervention, the same elements were the focus of attention.

What has been neglected thus far is the impact of the intervention on other outcomes

that are not explicitly addressed in the mainstream TQM literature but which are

nonetheless important outcomes. These outcomes include organizational commitment

and trust in management (used as a rough proxy for employee-management relations).

"Employee involvement is fundamental to TQM ideas" (Wilkinson, 1994, p279).

Employee involvement (EI) has been presented as a broad concept (Guest et al., 1993)

and includes activities directed at increasing communication to employees and

providing them with the opportunity to contribute to decision making in their work

259

place thereby increasing their commitment to the organization (Marchington et al.,

1994). 9 Guest (1992a) categorizes EI initiatives as follows: increased information to

employees; increased information from employees; changes to the design of work;

changes in incentives and; a more participative style of managing. Far removed from

the notion of Industrial Democracy, employee involvement in the context of TQM takes

the form of direct involvement in quality and improvement. This may take the form of

greater responsibility for quality in their job, contributions to improvements in their

work area and may involve a restructuring of work along the lines of semi autonomous

work groups.

Previously, the impact of the intervention was assessed in terms of the core elements of

TQM. This present assessment takes on board a different position; that is, viewing

TQM as representing one form of EI. Consequently, this view focuses on its impact on

EI outcomes, namely, organizational commitment and to a lesser extent, trust in

management. There is considerable debate and contradictory evidence for the

hypothesized EI-organizational commitment link (see Walton, 1985; Ahlbrant et al.,

1992 and Grummitt, 1983 for support for the link and Kelly and Kelly, 1991 and; Guest

et al., 1993 for contrary evidence). Marchington et al. (1994) provide positive and

negative evidence for EI and warn that EI is " as much affected by the prevailing

organizational culture as it is a source of change" (p890).

In order to assess the impact of the intervention on EI outcomes, a simple analysis (t-

tests) is conducted on participants and non participants in the intervention. The key

question being addressed is whether employee participation in the intervention has any

effect on organizational commitment and trust in management. This analysis was

restricted to the employee sample at site 1 and Table 11.3 presents the results.

9 Marchington et al. (1994) include the following as representing EI: house journals and employee reports, suggestion schemes, team briefings, Quality Circles, works councils and profit sharing.

260

Table 11.3: Independent t-tests between participants and non participants

Time 1 Time 2 Change over time


5.40 5.25

5.74*** 5.25

.34***

.00 Participants (n=87) Non Participants (n=78)

Trust in management

4.23 4.12

4.63** 4.24

.39**

.11 Participants Non Participants

*** significant differences between the two groups at .01 level ** significant differences between the two groups at .05 level

The results show that participants and non participants were not significantly different

prior to the intervention in terms of their commitment to the organization and their trust

in management. Significant differences were detected at time 2 and in the change that

occurred (T2-T1) between participants and non participants with participants being

more committed to the organization and having greater trust in management. In

addition, employees were asked a number of questions relating to the impact of the

intervention on management-employee relations. Significant differences were found

between participants and non participants in their overall mean score of the three

items 10 measuring the impact of the intervention on management-employee relations.

Participants were significantly more positive (mean score 4.83) than non participants

(mean score 3.4) on this dimension.

I° The particular items were as follows. The intervention has: improved communications between management and employees resulted in better relations between management and employees resulted in greater teamwork between management and employees

261

The results here would seem to support the EI-organization commitment link. As

previously argued, care must be exercised in comparing outcomes of particular EI

initiatives given that they can vary dramatically between organizations (Marchington et

al., 1994). The organizational context in which they are implemented may also

significantly affect the outcomes.

Taking TQM as one form of employee involvement, does it have an impact on

generally agreed upon EI outcomes? There is clear evidence that the intervention has a

significant effect on organizational commitment at both the individual and system

levels. Trust in management has significantly shifted in a positive direction between

time 1 and time 2 (results are not shown) for the overall group of employees. It is

plausible that the intervention may affect trust in management in a similar manner to

that found for management commitment to quality Overall, the evidence" suggests

that the TQM intervention has a significant impact on EI outcomes.

11.2.4 Goal achievement

This criterion assesses the intervention in terms of the extent to which the goals of the

intervention were achieved. As such it is consistent with evaluative research, "the

determination.... of the results.... attained by some activity... designed to accomplish some

valued goal or objective" (Suchman, 1967, p31-32). Among other objectives, Brooks

(1971) views evaluation as the determination of the extent to which the program

achieves its goals. This particular approach to evaluation has its strengths and

II While the evidence for organizational commitment is clear cut, it should be noted that the significant differences found in the t-tests (trust in management and an improvement in management-employee relations as a result of the intervention) do not control for the effects of any other variables.

262

limitations (see Schulberg and Baker, 1971) and rests on two fundamental assumptions

(not challenged here) that the goals are known and they can be measured.

Prior to proceeding with assessing the intervention in terms of goal achievement, one

issue needs to be addressed. How does this criterion differ from examining the

intervention in terms of its impact on the core elements of TQM? As some of the goals

of the intervention (team orientation and commitment to improvement) are the same,

these are not examined here. Rather the focus here is on the additional explicit goals of

the intervention that have not been previously examined.

The explicit remaining goals of the intervention include a change in managerial style,

greater employee involvement and an improvement in internal customer-supplier

relations. In order to assess the extent to which the intervention achieved these goals,

two types of evidence are employed. First, goal achievement is assess by examining

the extent of change that has occurred over time. This involves a shift from assessing

the impact of the intervention at the individual level to a subsystem level; that is,

supervisors as a group and / or employees as a group. Second, a more subjective

method of goal achievement is included which consists of management's view of what

the intervention has achieved.

The overarching objective of the intervention was to change the culture of the

organization to one of participative involvement to achieve continuous improvement

and teamwork. Within this broad objective, a primary focus of the intervention was on

the style of managing. In particular, the training and education concentrated on Theory

X and Y views of employees, notions of empowerment and leadership styles with

specific emphasis on participative style.

263

To assess if there has been a change in managerial style or the thinking behind it, the

dimensions used are: Theory X and Y, participative style and supervisory commitment

to quality. In terms of supervisory subscription to Theory X (4.02 at time 1 and 3.87 at

time 2) there has been a shift in a negative direction although it is not significant.

However, there has been a significant positive shift in Theory Y (5.08 at time 1 and

5.39 at time 2). Therefore, it would seem that there has been some shift in managerial

thinking Does this translate into a shift in behaviour? Initially, looking at supervisory

perceptions of the behaviour of their immediate superior, perceptions of the

participative style of their immediate superior has not significantly changed (4.95 at

time 1 and 5.02 at time 2). This lack of significant change is also found in perceptions

of commitment to quality (4.87 at time 1 and 5.01 at time 2). A contrasting picture

emerges if one looks at employee perceptions. Employee perceptions of supervisory

participative style has significantly shifted in a positive direction between time 1 and

time 2 (4.98 at time 1 and 5.15 at time 2). In terms of supervisory commitment to

quality, there has been a positive shift although not significant. However, employee

perceptions of management commitment to quality has significantly shifted in a

positive direction (5.10 at time 1 and 5.27 at time 2).

Has the goal of managerial style been achieved? The evidence suggests that there has

been a significant change in supervisory subscription to Theory Y as well as employee

perceptions of the participative style of their immediate boss. However, in contrast,

supervisory perceptions of the behaviour of their immediate superiors (participative

style and commitment to quality) has not significantly changed. If we take

achievement to mean significant change, then the goal of change in managerial style

has been achieved at the employee level but not at the supervisory level despite the

significant shift in Theory Y. Relaxing the notion of goal achievement and asking if

264

progress has been made toward that goal, the evidence does suggest that progress has

been made.

Greater employee involvement in the process of continuous improvement was another

objective of the intervention. This is a broad and imprecise thus difficult to measure

goal. One interpretation of this goal could be the number of employees participating in

the intervention. Another view may see greater employee involvement in term of

employees having greater influence over decisions in their work and / or work area. In

dealing with the first interpretation, 52.7% of employees reported that they were

participating in the intervention. What does this mean in relation to the stated goal? Of

course, 52.7% of employees participating in the intervention is better than 30% but not

as good as 70%. If one takes the goal of 'total employee involvement', then the goal

has not been achieved although visible progress toward that goal is evident.

Has there been a reduction between desired and actual influence for the group of

employees? The results of paired t-tests indicate that the discrepancy between desired

and actual influence has remained stable over time (0.84 at time 1 and 0.83 at time 2)

for employees. Thus, at a system level, there has not been a significant change in terms

of a reduction in the influence gap between desired and perceived actual influence.

If the goal of employee involvement is viewed as eliciting all employees' participation

in the intervention or allowing them greater influence over work related decisions, the

goal has not been achieved.

The final objective of the intervention was to improve internal customer-supplier

relations; relations between work areas / departments. While this goal was not

measured directly, two proxy measures exist; satisfaction with interdepartmental

265

relations 12 and trust in other departments 13 . An improvement in relations between

departments should lead to greater satisfaction with interdepartmental relations and

greater trust in other departments. As supervisors / managers are more proximal to

other departments than employees, in theory, one should see an improvement initially

at this level of the hierarchy. Supervisors do not report greater satisfaction with

interdepartmental relations (5.11 at time 1 and 5.28 at time 2) nor is there a significant

improvement in trust in other departments (4.53 at time 1 and 4.58 at time 2). A similar

picture is provided by employees 14 .

This lack of significant positive change in interdepartmental relations is consistent with

the state of progress of the intervention. At the time of the second round questionnaire,

the issue of customer-supplier relations was just being confronted. This involved an

internal customer- supplier audit 15 which it was hoped would provide a mechanism for

improving customer-supplier relations. This had commenced at the time of the second

round questionnaire.

To conclude, in simple terms, the absolute goals of the intervention have not been

achieved. However, one could argue that this is a harsh approach to adopt in view of

the short time span allowed for the achievement of goals. Consequently, rather than

focusing on the achievement of goals, it may be more realistic to assess the intervention

12 This was measure with the following items: Satisfaction with the support of my group gets from other departments Satisfaction with the opportunities to discuss matter with other departments Satisfaction with the support my depai tment gives to other departments

13 This was measured with the following items (strongly agree - strongly disagree) Other departments can be relied upon to do as they say they will do Other departments put the overall organization's benefit before their own benefit Overall, there is a lot cooperation between departments

14 Satisfaction with interdepartmental relations 4.51 at time 1 and 4.56 at time 2. Trust in other depai tinents 4.47 at time 1 and 4.42 at time 2. 15 • • This involves representatives from two work areas/ departments coming together to discuss the outputs of one depai anent for another; what are the important outputs, what can be improved and how will the important outputs be measured (to act as a monitoring device)

266

in terms of progress toward those goals or an improvement compared to the situation

prior to the intervention. Adopting this approach, there is evidence of improvement and

progress toward those goals. This view that some progress has been made is consistent

with the views of senior level managers at the site on the overall impact of the

intervention. While they recognized that some progress has been made, they were

disappointed at the speed at which change was happening.

11.2.5 A TQM view

So far, the intervention has been assessed by: its effect on the core elements of TQM;

its impact on performance improvement; the degree to which EI outcomes were

affected and; the extent to which it achieved its goals. The final criterion (if it may be

called this) is to assess the intervention from a TQM perspective; how would TQM

proponents assess the impact of the intervention?

Achieving culture change is central to what TQM is about (Wilkinson, 1994; Hill,

1995; Oakland, 1989). From what is known about culture change (Schein, 1985) years

rather than months is the appropriate time frame. Blackburn and Rosen (1993) in their

study of Baldrige Award winning companies report that "none of the firms made these

changes overnight. Individuals talked in terms of evolution not revolution; years and

not weeks of effort" (p62). This, in conjunction with the long term emphasis

underlying TQM would provide TQM proponents with grounds to argue that the

benefits would be in the longer term. Hill and Wilkinson (1995) state that a period of

up to seven years for TQM to become the way of managing would not be unusual.

Whyte and Witcher (1992) argue "that TQM is after all a radical new approach to

business management. The true benefits will be in the longer term " (131).

267

If the benefits materialise in the longer term, what, if anything, does this mean for the

change that occurs or needs to occur? Reger et al. (1994) cite a study by Ernst &

Young and the American Quality Foundation (1992) which recommended a gradual

implementation of TQM to more synoptic efforts. The authors in question argue that

TQM programs that are presented as "radical departures from the organization's past

fail because the cognitive structures of members.... constrain their understanding and

support of new initiatives" (p566). Rather the authors argue for a step by step

introduction of a moderate degree of change.

The short term nature of this evaluation would be perceived as antithetical to the long

term emphasis underlying the TQM philosophy. Consequently, proponents of TQM

would argue that the impact of TQM should be evaluated on a longer term basis.

However, from a TQM stance, the impact of the intervention may be viewed as a

positive start in the implementation of TQM that needs to be reinforced and built upon

with other changes.

11.3 Discussion

In examining the impact of the intervention on the core elements of TQM, the results

suggest that while the intervention had an impact at the individual level, this did not

lead to an overall improvement in the system. Therefore, in principle, the TQM

intervention did have an effect but in the short term, the intervention was ineffective in

leading to overall improvements. In terms of perceived performance improvement, the

intervention was found to have a significant impact at the individual level. In

particular, the behavioural components of the intervention; employee participation and

supervisory reinforcement were found to be better predictors of perceived performance

268

improvement than the key elements (in this study) of TQM. This highlights the

importance of changing behaviour as a route to performance improvement, at least in

the short tenth

Assessing the impact of the intervention against its goals, the evidence suggests that the

goals have not been achieved although progress has been made. A more favourable

assessment emerges if the impact of the intervention is measured against EI outcomes.

Finally, an attempt was made to assess the impact of the intervention from a TQM

perspective. Having assessed the intervention using five different criteria, what overall

conclusion can be drawn?

Overall, there is evidence to suggest that in the short term, the intervention has had a

minimal effect in terms of overall significant improvement at the system level. Against

this unfavourable assessment is the positive impact of the intervention on EI outcomes.

Does this detract from the previous evidence? It would seem that the evidence on the

side of the intervention having a minimal impact holds ground despite the positive

effect on organizational commitment. Why would this be the case? First, it is possible

that the positive effect on EI outcomes is an unintended consequence of the

intervention. Second, organizational commitment may be more easily affected

compared to outcomes more fundamental to TQM. What this conclusion does is

neglect the time span permitted for effects to take place.

Given the short time span allowed for the assessment, what change could one expect as

a result of the TQM intervention? There exists little empirical work that focuses on the

process of change or rather the implementation of TQM. One key finding of Hill's

(1995) research on the implementation of TQM was the time it took to cascade

throughout the organization. Even in the organization which had the most explicit view

269

of TQM and a comparatively more prepared scheme, implementation began with top

management and took two years before it reached the lowest level in the hierarchy. A

similar picture is painted of the implementation of TQM in the remaining organizations.

With this type of time frame as a guideline for the implementation of TQM, one may be

inclined to argue that in the initial period, the overall extent of change may not be that

significant. However, as a counterargument, one would expect to see significant

change in the group of managers / supervisors given that this group would not only

have greater but also longer experience with the intervention.

This chapter highlights several issues. Given the relatively early stage in the

development (both theoretically and empirically) of TQM, the use of multiple criteria

of assessment is necessary in order to fully evaluate the impact of TQM interventions.

By doing this, it was shown that while the present intervention had minimal impact on

the key elements of TQM, a stronger effect was found for EI outcomes. Second, a

distinction needs to be made between impact at the individual level and an overall

improvement in the system. With this distinction in mind, the intervention had a

significant effect at the individual level on all the elements (in this study) of TQM.

However, an overall improvement in the system was not evident. Finally, several

questions need to be addressed regarding change. Could there have been greater

change? Why was more change not visible? The concluding chapter picks up these

questions.

11.4 Conclusions

This chapter set out to make an overall assessment of the impact of the intervention. To

do this, the impact of the intervention was assessed using the following criteria: its

impact on some of the core elements of TQM; the degree to which the intervention

270

influenced perceived performance improvement; its effect on employee involvement

outcomes; the extent to which the goals of the intervention were achieved and finally;

an assessment from a TQM perspective.

Overall, the results indicate that the intervention was not effective in leading to an

overall positive change in the key elements of TQM. This may be partly due to the

short time allowed for the effects of the intervention to occur. It may also be due to the

nature of the intervention. The concluding chapter addresses the following two

questions: could there have been greater change? And why wasn't there?

271

Chapter 12: Conclusions

12.1 Introduction

Adopting a number of different criteria of assessment, the prior chapter assessed the

impact of the intervention. The evidence suggests that at the individual level, the

intervention was found to have a significant total effect on all the key elements in this

study. Of these elements, the intervention was found to have a significant direct effect

only on improvement as part of the job. Thus, if one relied exclusively on assessing the

impact of the intervention on the basis of its direct effects, one would have concluded

that the intervention has a significant impact solely on improvement as part of the job.

By including the indirect effects, a different conclusion emerges; that is, the

intervention has a significant total effect on all the elements.

The intervention did not lead to an overall significant improvement in the key elements

of TQM. This conservative conclusion ignores the significant improvement in the two

dimensions of commitment to improvement based on its retrospective measurement at

time 1. In assessing the impact of the intervention in terms of goal achievement, the

conclusion drawn was that progress had been made toward achieving the goals of the

intervention. This, in addition to the significant total effect of the intervention at the

individual level and a long term perspective underlying TQM, would suggest that

possibly greater time is needed for the positive effects to materialise. Finally, a more

positive assessment emerges from assessing the intervention from an Employee

Involvement (EI) perspective. This assessment highlights the unintended positive

consequences of the intervention.

272

The conclusion that the intervention had a minimal overall impact raises several issues

concerning the content and process of change. First, to what extent are the results

found a function of the time span allowed for the effects to occur? In other words, are

the results primarily due to the time lag rather than the content and process of change.

Putting aside the time lag as a potential explanation of the results, two further questions

are raised. First, why did greater change not occur? And could greater change have

occurred? These issues have potential implications not only for the content of TQM

interventions but also the way they are implemented.

This concluding chapter begins by discussing the key findings of this study and their

implications. Then, the questions concerning change are addressed. The penultimate

section of this chapter deals with the broad and specific limitations of this research.

This provides a basis for outlining directions for future research and areas for

subsequent development.

12.2 Main findings

A key finding of this research, at a theoretical level, concerns the conceptual

independence of intrinsic motivation and commitment to improvement. Overall, in the

TQM literature, several concepts such as commitment to quality, commitment to

improvement and intrinsic motivation are widely used but not differentiated. To my

knowledge, there has been little attempt to operationalize commitment to improvement.

Consequently, a definition of commitment to improvement was put forward in this

thesis. Reasonably strong evidence was provided for treating commitment to

improvement not as a unidimensional construct but as consisting of two dimensions:

general orientation to quality and improvement as part of the job. Furthermore, support

273

was found for treating the two dimensions of commitment to improvement and intrinsic

motivation as independent constructs.

Second, the findings suggest that the intervention affected team orientation, general

orientation to quality, improvement as part of the job and intrinsic motivation in a

different manner For example, supervisory reinforcement of the intervention has a

significant (total) effect on team orientation. Perceived benefit of the intervention has a

significant (total) impact on intrinsic motivation and a significant direct effect on

improvement as part of the job. Finally, perceived appropriateness of the intervention

has a significant (total) effect on general orientation to quality. This suggests that the

intervention has a differential effect on the outcomes in two ways. First, the

intervention affects the outcomes differently in terms of direct and indirect effects.

Second, different aspects of the intervention affect the outcomes in a different manner

Therefore, the practical implications seem to be clear cut. It would seem that if injob'

attitudes (intrinsic motivation and improvement as part of the job) are to be affected, it

is important that individuals perceive the intervention as providing benefit to them. On

the other hand, if the key target of change is òutjob' attitudes (team orientation and

general orientation to quality); that is, toward the work area and colleagues, the results

suggest that aspects other than perceived benefit are important. Organizations

implementing TQM will want to change both ìnjob' and òutjob' attitudes. Therefore,

the intervention needs to be assessed by organizational members in a favourable light

along a range of dimensions. An exclusive emphasis on portraying the intervention as

beneficial, if these results are broadly correct, will have no subsequent impact on

òutjob' attitudes. Consequently, organizations need to take steps to increase the

likelihood that the intervention is perceived not only as beneficial but also as

appropriate and that it is reinforced by the immediate supervisor.

274

How an individual judges and assesses the intervention is important in affecting

subsequent change This leads to the third main finding of this study concerning the

relative effect of employee participation in and assessment of the intervention. Overall,

the results suggest that it is the assessment of the intervention that is a better predictor

of subsequent changes rather than participation in the intervention per se, which was

not found to have a significant effect on any of the outcome variables. Therefore,

involving employees in the TQM intervention is not a sufficient condition to affect

change. What is of greater significance is how employees perceive and assess the

intervention.

The lack of predictive power of participation on subsequent attitudes deserve attention

given its emphasis in the Quality Circle (QC) and Employee Involvement (EI)

literature. Several explanations were put forward for this finding. Participation may

not have matched pre participation expectations thus having no effect or a damping

effect on subsequent attitudes. Second, much of the literature on TQM, QC and EI

programs stresses the voluntary nature of employee participation. However, in practice,

little is known about the extent to which individuals are put under pressure from peers

and/or their immediate boss to participate. Thus, the extent to which an individual is

put under pressure to participate may influence the lack of change in subsequent

attitudes.

Therefore, the results highlight the importance of tapping an individual's assessment of

the participatory mechanism as an important explanatory factor of the outcomes. From

this, it is suggested that the emphasis should move away from a direct participation-

outcome linkage to a participation-assessment-outcome linkage. By including

275

individuals' assessment of what they are participating in may shed light on why (or

why not) participation does (or does not) lead to desired outcomes.

More specifically, this finding has policy implications for the implementation of TQM

interventions. For affecting change, it is not sufficient to involve employees in the

intervention per se. What the results suggest is that in the design and the subsequent

implementation of TQM interventions, emphasis needs to be placed on how the

intervention will be perceived by employees. As discussed in chapter 1, considerable

emphasis is placed on 'total participation' in the TQM literature. Following from this,

the prescription for organizations would emphasize getting everyone in the organization

participating in the intervention. The results here suggest that an equal, if not more

important consideration, is how the intervention is judged by organizational members.

The latter consideration cannot be subordinate to the first; total participation (if this is

realistic) with a less than favourable assessment may not lead to the desired outcomes

in terms of attitudinal change.

The fourth finding relates to supervisory behaviour. A consistent finding throughout

the results is the importance of supervisory behaviour in affecting employee attitudes

and behaviour. Perceived supervisory behaviour was found not only to have an effect

on the key elements (in this study) of TQM but also on employee participation in and

assessment of the intervention. Where it was tested, the results indicate that the

immediate supervisor has significantly greater effect than management as a group on

employee attitudes and behaviour.

This finding highlights the importance of the immediate supervisor which seems to be

overlooked in the TQM literature. An overwhelming and consistent emphasis is placed

on the commitment of senior management to TQM and management commitment to

276

quality as a vital ingredient for success in TQM efforts. In contrast, far less importance

is assigned to leadership down the hierarchy as evidenced by the general absence of

discussion on the role of front line supervisors. There are two possible explanations for

this lack of emphasis given to supervisors. First, their role may be deemed

comparatively unimportant in contrast to senior management. Second, there may be an

implicit assumption that the commitment of senior management will cascade smoothly

to first level supervisors.

What the findings here suggest is that at the employee level, the important influence

comes from the immediate supervisor. This does not negate the importance of senior

management commitment but indicates that in itself it does not directly affect employee

attitudes and behaviour. The inherent chain of example setting of the level above for

the level below can be broken at any stage. Consequently, while senior management

may be committed to the philosophy of TQM, this must be cascaded down to the lowest

supervisory level so that employees, in turn, are affected. Thus, while commitment

from senior management is a necessary condition for the success of TQM, it is not a

sufficient condition on its own to instill quality oriented attitudes and behaviour in

employees.

Finally, organizational change interventions do not occur in a vacuum; that is, prior

experience, attitudes and behaviour of organizational members have an effect on how

the intervention is assessed and subsequently how successful the intervention is in

terms of changing attitudes and behaviours. For example, an individual's prior

commitment to the organization and previous experience of a participative program

prior to the intervention influences how they assess the intervention in terms of

perceived benefit. This has implications for evaluation studies and for organizations

implementing change initiatives such as TQM.

277

In terms of evaluating organizational change interventions, care needs to be exercised

in the attribution of cause and effect. Organizational commitment, for example, has

been used as an outcome of employee involvement programs and more broadly of

HRM practices and policies. Evaluating the impact of these interventions in terms of

the effect on organizational commitment ignores the potential effect of organizational

commitment on the change intervention in the initial instance. Therefore, what is

presented is an oversimplified picture of the process by which change occurs. The

results here suggest that the process of organizational change is more complex;

organizational commitment was found to be an antecedent of how individuals assessed

the intervention and also an outcome of the intervention. Therefore, to treat

organizational commitment solely as an outcome may misrepresent how change occurs

in organizations.

The policy implications for organizations implementing TQM interventions are as

follows. First, individuals' prior experience of employee involvement activities do

influence their future assessment of similar activities. Therefore, if an organization's

previous involvement activity was largely unsuccessful and it is embarking upon a

TQM intervention it may be necessary from the onset to ensure that the TQM

intervention is disassociated and clearly differentiated from the previous activity

especially if the time lapse is reasonably short. What may be required pre intervention

is the creation of a state of readiness to change, similar to Lewin's (1951) unfreezing so

as to increase the likelihood of attitudinal change occurring. As mentioned in chapter

2, given the previous changes (prior to the intervention) at the site which contributed to

the site's transformation from a loss making situation to one of profitability, it is quite

likely that with the intervention, there was an overall lack of perceived need for change

at the site.

278

Second, an individual's attitude toward the organization prior to the intervention has an

effect on how the intervention is assessed. Clearly, this presents a dilemma if attitudes

prior to the intervention affect how the intervention is assessed which in turn affects the

degree to which attitudinal change occurs. In hindsight, greater attention to creating a

stimulus for change may have facilitated greater change.

Finally, as indicated at the beginning of this chapter, the TQM intervention had a

minimal overall impact on attitudes. This provides the grounding for the ensuing

discussion of why greater change was not evidenced at the site.

12.3 The extent of change

Overall, there was little change evidenced between time 1 and time 2. What, if any,

factors may help explain this apparent stability in attitudes and behaviours that were the

focus of the training and education. Several possible contributory factors to the lack of

change are discussed. The first three factors relate to the intervention: its content,

reinforcement and transferability, and the process of change. The final possible factor

is the time span allowed for the effects to take place.

In terms of the content of the TQM intervention, the training and education program

constituted the foundation of and signified the start of the TQM intervention. It was

categorized as a 'soft' intervention due to its emphasis on attitudes, values and

indirectly behaviour. Regarding change, two assumptions were in operation. First,

change at the level of employees is dependent upon change occurring at the managerial

level. Second, the education and training program would affect attitudinal change

which in turn would affect behaviour.

279

The content of the training and education program is consistent with TQM. Due to a

lack of available descriptive accounts of other TQM interventions, it is not possible to

ascertain how this intervention matched others regarding content. Whether a different

training content would have induced greater change is an open question. However, a

more damaging factor is the underlying assumption regarding change; that is, that

participation in the training program would lead to attitudinal change which

subsequently would result in behavioural change.

The primarily indirect approach to changing behaviour via attitudes, as previous

research indicates, may not be as successful as directly focusing on behavioural change.

Therefore, using a training and education program as the primary driver of change may

not produce the desired results. Guest and Peccei (1994) argue that interventions using

training as the initial lever of change have a history of failure. Consequently, the lack

of change resulting from an intervention relying heavily on education and training is

not surprising in light of the previous research findings. Therefore, organizational

change interventions may increase their chance of affecting change by adopting a dual

focus on both attitudes and behaviour. At this site, as a result of the perceived lack of

progress occurring, at the end of this evaluation, targeting behaviour via the

performance objectives and appraisal of managers was now the focus of the change

efforts. Had this been implemented at the beginning to reinforce the training and

education, possibly, greater change may have occurred in the short term.

The second potential contributing factor to the lack of change is a lack of reinforcement

of what was learned on the training and education program. As mentioned in chapter 2,

considerable preparation was given to the design of the intervention and how the

training was to be cascaded down the managerial hierarchy. Therefore, while all

280

managers and supervisors participated in the training program, when they returned to

the 'workplace' and organizational reality, the 'environment' did not support or

reinforce the change. In other words, applying the principles of learning in training

back in the work environment was problematic.

The issue of reinforcement is vital to increasing the chances of training programs

inducing change back in the organization. One potentially powerful source of

reinforcement is the behaviour of the immediate superior. This type of reinforcement

was not strong within the managerial hierarchy as evidenced by the lack of significant

difference in perceptions between supervisors and employees in terms of the extent to

which their immediate boss reinforced the intervention. In addition, other mechanisms

such as performance appraisal were not in place to stimulate the transferability of the

training. Consequently, the lack of reinforcement may have contributed to the

ineffectiveness of the training program.

A related factor is the process of change. The onus and responsibility was on the

individual supervisors to cascade the training to their employees. If reinforcement of

the training was generally weak, as in the case of supervisors, this would have a knock

on effect in the cascading of the training to employees. Thus relying on supervisors to

cascade the training was detrimental given that the training was not reinforced at the

supervisory levels.

To summarize, the heavy reliance on training and education without strong

reinforcements back in the organization seem to have reduced the potential effect on, in

the first instance, change within the supervisory hierarchy and consequently, change at

the bottom of the organization.

281

The time span allowed for this evaluation study may have had a strong effect on the

findings; that is, the general lack of change. One could argue that in this case, given the

process by which change was planned (i.e. the cascading process), an adequate time

frame was not adopted. Supervisors / managers were returned to face organizational

reality and the associated pressures of their job, consequently, immediate action in

terms of cascading the intervention may not have been feasible or realistic. Even if the

intervention was perceived as a high priority and reinforced through organizational

mechanisms, the very nature of cascading a change process takes time.

The potential influence of the time lag on the results has been raised but needs to be

pursued further. Specifically, to what extent can the time lag account for the results?

Greater reinforcement of the intervention may have speeded up the cascading process.

Consequently, while the time lag may be a factor, attention needs to be placed on the

process of change. Second, if the time lag was a major factor, one would expect to see

a significant change in the attitudes, perceptions and behaviours in the target subsystem

(first order change) of supervisors/managers. If this were the case, one would be in a

stronger position to argue that a greater time lag is needed for effects to occur.

Pursuing this further by looking at second order change; that is, change in the

subsystem of employee participants in the intervention, the evidence suggests that

minimal change has occurred.

Therefore, while the issue of the time lag may be a factor in explaining the overall lack

of significant change, it may not be the most important factor. The argument presented

here was that if the issue of time per se was a major factor, one would expect to see

significant change in the subsystem proximal to the intervention. In this case, this has

not happened, therefore, other factors need to be taken into account to explain the

extent of change that occurred.

282

12.4 Greater change?

Could greater change have occurred? In other words, taking the content of the

intervention as given, what other steps could have been taken to increase the likelihood

of greater change occurring?

There was a lack of visible planning as to what would happen once all supervisory

personnel had completed the training program. Too much attention and energy was

invested in the design and cascading of the training within the supervisory ranks at the

expense of what was to happen post training Probably, what was needed at the

preparation stage was a systems view of training. Viewing the training as a subsystem

within an overall system may have indicated the changes that needed to occur in other

systems so that all the subsystems were reinforcing each other. This may have

highlighted the need to modify performance objectives and appraisal to reinforce the

change and highlight the importance attached to the intervention.

If other organizational mechanisms had been used from the onset to reinforce the

content of the intervention, greater change may have occurred. This may have reduced

the uneven cascading that occurred and gone some way toward building a momentum

for change within the managerial ranks prior to involving the shopfloor in the

intervention. In addition greater reinforcement may have reduced the length of time it

took to cascade the intervention. Simply, rather than relying primarily on affecting

attitudes as a way to change behaviour, a dual focus on attitudes and behaviour may

have resulted in greater change.

Finally, as mentioned previously, at the preparation stage, greater attention on creating

a stimulus for change may have helped. Considering the previous changes that had

283

been implemented, what was the perceived motivation to embark upon further changes?

The necessity for change may not have been perceived as such by organizational

members. Survival was the key stimulus underlying the previous major change in the

reorganization of production. This stark impetus was absent in relation to the

intervention.

12.5 Limitations of the research

Two limitations have been previously discussed: the lack of a priori control group and

the short time span allowed for the change to occur. The early intervention in the

change process was necessary in order to establish a post hoc control group. However,

this time span is quite common in terms of the initial post intervention measurement.

This study focused on the implementation of a 'soft' TQM intervention and thus the

findings are limited to this type of intervention. However, this intervention and its

reliance on training and education typifies what is prescribed by the quality advocates.

It would have been interesting to include another TQM intervention that differed in

terms of its content.

The more specific limitations of this study are as follows. First, the focus was on two

of the three elements portrayed as central to TQM. The notion of customer satisfaction

was not included in this study. Second, it would have been useful to have collected

more detailed qualitative data regarding the process of change. In particular, data on

pre intervention expectations may have shed light on the subsequent effects of the

intervention. Furthermore, a systematic approach to interviewing the managers /

supervisors post intervention may have provided greater insight into the uneven

cascading process and their interpretation of what was needed in terms of making the

284

intervention successful in affecting change at the site. In other words, what were the

perceived obstacles to change and where were they?

Another limitation of this study is the use of perceived performance improvement.

Ideally, objective performance measures would have been collected independently at

time 1 and time 2. This would have allowed a more rigorous test of the links between

behaviour, attitudes and performance in the context of TQM. Similarly, the use of a

retrospective measure of the two dimensions of commitment to improvement at time 1

is clearly not ideal. Finally, the measure of perceived benefit of the intervention is

crude in that it does not tap the source of benefits; that is, what individuals perceive the

benefits to be.

12.6 Future research

With a relatively new organizational phenomenon such as TQM and its multifaceted

nature, research possibilities abound. Consequently, the potential future directions for

research are discussed in the context of this study. First, the methodological issues are

discussed followed by future avenues for theoretical development.

The most important and most difficult direction for future investigation is longer term

evaluations of TQM interventions. To my knowledge, a systematic long term

evaluation of TQM has not yet been conducted. This would provide a more appropriate

time span to evaluate the effects of a TQM intervention. In this study, it would be

interesting to evaluate the subsequent progress of TQM in light of the changes that

were beginning to happen when this evaluation ended. In addition, a longer term

evaluation would provide rich detail as to the process of change and provide useful

insights into whether short term effects result in more enduring changes.

285

A second methodological consideration involves the use of a control site that is

comparable on all important respects except for the TQM intervention. As previously

mentioned, this evaluation was of a 'soft' TQM intervention, consequently, future

research may want to consider a comparison of the effects of different types of TQM

interventions.

The third methodological issue concerns endogeneity in evaluation of change

interventions. Future evaluation research should provide a more comprehensive and

detailed assessment of the process of change. This means that due consideration needs

to be given to the effect of prior attitudes held by individuals on how they interpret and

assess a particular change intervention. Subsequently, the effects of the change

intervention may be more accurately reflected.

The first area for subsequent theoretical development is the models containing the

antecedents of team orientation and commitment to improvement. The models most

likely do not include all the relevant and important antecedents of team orientation and

commitment to improvement. For example, it would be better to tap individuals'

perceptions directly of the extent to which management exhibited team orientation and

reinforced team orientation at lower levels rather than indirectly tapping this via

management commitment to quality as was done here. Similarly, in the commitment to

improvement model, an important antecedent may be the importance attached to

continuous improvement within the work group. Future research should include deeper

theoretical development on the antecedents of these important outcomes of TQM.

A related area for subsequent theoretical development is the conceptualization of some

of the constructs used here. Future research should explore the dimensionality of the

286

concepts and develop more comprehensive measurement instruments. Commitment to

improvement and intrinsic motivation were found to be independent constructs in this

study. However, there was no attempt made to measure commitment to quality and

assess its conceptual independence from these two concepts.

The theoretical links between the TQM intervention and the intervening variables

warrant future attention. Specifically, what mechanisms explain the effect, for

example, of perceived supervisory reinforcement of the intervention on higher order

need strength. Similarly, other links between the intervention and the intervening

variables warrant explicit theoretical grounding.

Finally, as an organizational change intervention, the underlying theoretical basis of

TQM needs to be developed in two related directions. First, there is little theory to

explain why some TQM interventions are successful and others not. Therefore, one

theoretical line of development would be the identification of factors that distinguish

successful and unsuccessful interventions. Second, it is easy to attribute the failure of

TQM interventions to faulty implementation rather than challenging the validity of the

theoretical basis and assumptions of TQM. Here, empirical research and findings from

other disciplines may make a contribution by developing a stronger theoretical basis

thereby providing a better guide to the implementation of TQM for practitioners.

287

APPENDIX 1: Questionnaires

Employee 1st round questionnaire

Supervisor 1st round questionnaire

Employee 2nd round questionnaire

Supervisor 2nd round questionnaire

288

EMPLOYEE 1sT ROUND QUESTIONNAIRE

The London School of Economics and Political Science

THE INFORMATION CONTAINED IN THIS QUESTIONNAIRE SHALL REMAIN COMPLETELY CONFIDENTIAL. NO ONE WITHIN WILL SEE ANY OF YOUR RESPONSES.

SECTION 1 What is your job title? (Please circle one number)

Module Operator 1 Purchase Progress Controller 5 Module Craftsman 2 Analyst 6 Materials Controller 3 Buyer 7 Engineer 4 Clerical 8

Other

What department/module are you in? (Please circle one number)

Finance & Planning 1 Alternators 8 Engineering 2 FVE/CSG 9 Quality 3 CA45 10 Market Development 4 Other Starters 11 Customer Interface 5 Process 12 Service 6 Thermostat 13 Inline 7 Supplies 14

Within your Module/Department, what cell or section are you in?

Not appropriate 0 Cell/Section Name/No

How long have you been in your present job?

What year did you join ?

What year did you join the division of

How old are you?

Are you male or female? Male 0 Female 1

289

SECTION 2 REGARDING YOUR PRESENT JOB, HOW SATISFIED ARE YOU WITH THE FOLLOWING: (Please circle the number that best reflects how you feel)

Extremely dissatisfied

Very dissatisfied

Moderately dissatisfied

Not sure Moderately satisfied

Very satisfied

Extremely satisfied

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

7 1 2 3 4 5 6

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

Your physical work conditions

The freedom to choose your own method of working

Your fellow workers

The recognition you get for good work

Your immediate boss

The amount of responsibility you are given

Your rate of pay

The degree to which you are fairly paid for what you contribute to the organization

Your opportunity to use your abilities

Industrial relations between management and workers

Your chance of promotion

The way your firm is managed

The attention paid to suggestions you make

Your hours of work

The amount of variety in your job

Your job security

The amount of influence you have over day to day work decisions that affect you

290

The offer of a bit more money with another employer would not seriously make me think of changing my job

SECTION 3 PLEASE CIRCLE THE NUMBER THAT BEST SHOW HOW YOU FEEL Regarding the site:

Even if were not doing too well financially, I would be reluctant to change to another employer

To know my own work had made a contribution to the good of would please me

In my work I like to feel I am making some effort, not just for myself but for as well

I am quite proud to be able to tell people I work for

I feel myself to be part of

Strongly disagree

Disagree quite a lot

Disagree just a little

Not sure

1 2 3 4

1 2 3 4

1 2 3 4

1 2 3 4

1 2 3 4

1 2 3 4

Agree just a little

Agree quite a lot

Strongly agree

5 6 7

5 6 7

5 6 7

5 6 7

5 6 7

5 6 7

SECTION 4 REGARDING YOUR WORK, HOW IMPORTANT DO YOU THINK THE FOLLOWING ARE: (Please circle the most appropriate number)

Using your skills to the

Not at all important

Not particularly important

Not sure about its

importance

Moderately important

maximum 1 2 3 4

Achieving something you personally value 1 2 3 4

The opportunity to make your own decisions 1 2 3 4

Challenging work 1 2 3 4

Extending your range of abilities 1 2 3 4

Fairly important

Very important

Extremely important

5 6 7

5 6 7

5 6 7

5 6 7

5 6 7

291

SECTION 5 PLEASE CIRCLE THE NUMBER THAT BEST SHOWS HOW YOU THINK

Strongly disagree

Disagree quite a

lot


Not sure Agree just a little

Agree quite a lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

If I got into difficulties at work I know my workmates would try and help out

I can trust the people I work with to lend me a hand if I need it

I have full confidence in the skills of my workmates

Most of my workmates can be relied upon to do as they say they will do

I can rely on other workers not to make my job more difficult by careless work

Most of my fellow workers would get on with the job even if supervisors are not around

My workmates are genuinely committed to improving quality

My workmates have specific ideas about how to improve the quality of their work

SECTION 6 REGARDING THE FOLLOWING STATEMENTS, PLEASE CIRCLE THE NUMBER THAT BESTS SHOWS HOW YOU FEEL

Strongly disagree

Disagree quite a ' lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

292

I feel a sense of personal satisfaction when I do this job well

My opinion of myself goes down when I do this job badly

I take pride in doing my job as well I can

I feel unhappy when my work is not up to my usual standard

I like to look back on the day's work with a sense of a job well done

I try to think of ways of doing my job effectively

SECTION 7 PLEASE INDICATE WHAT YOU THINK ABOUT THE FOLLOWING STATEMENTS BY CIRCLING THE MOST APPROPRIATE NUMBER Regarding management at this site:

Strongly disagree

Disagree quite a

lot


Not Sure Agree just a little

Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

5 1 2 3 4 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

Management is sincere in its attempts to meet the workers' point of view

I feel confident that will always try to treat me fairly

Management would be quite prepared to gain advantage by deceiving the workers

Management at work seems to do an efficient job

Management can be trusted to make sensible decisions for the firm's future

Our division has a poor future unless it can attract better managers

Management is genuinely committed to improving quality

Management sets examples of quality performance in their daily activities

Management does its best to provide employees with the right tools and materials to do a quality job

Management has attempted to involve everyone in continuous improvement

Management provides support for quality improvements throughout the organization

293

SECTION 8 (Please circle the most appropriate number)

Overall, how satisfied are you with:

Very Dissatisfied Rather Neither Fairly Satisfied Very dissatisfied Dissatisfied satisfied nor satisfied satisfied

dissatisfied

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

as an employer

Quality of supervision

Your present job

The amount of training on quality you have received

The content of your training on quality

The support my group gets from other departments

The opportunities to discuss matters with other departments

The support my department gives to other departments

SECTION 9 Please describe the person you usually report to by circling the number that best reflects what you think:

The person I normally report to:

Strongly Disagree Slightly Neither Slightly Agree Strongly disagree disagree agree nor agree agree

disagree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

294

Is genuinely committed to improving quality

Encourages me to suggest improvements in the organization of my work

Gives me feedback on my suggestions for improvement

Gives me more recognition when I produce high quality work

Influences how I feel about quality

Gives priority to finishing work on time rather than the quality of work

Sets example of quality performance

Strongly disagree

Disagree Slightly disagree

Neither agree nor disagree

Slightly agree

Agree Strongly agree

in his/her day to day activities 1 2 3 4 5 6 7

Demands that people give their best effort 1 2 3 4 5 6 7

Insists that subordinates work hard 1 2 3 4 5 6 7

Gives me enough information to enable me to do a quality job 1 2 3 4 5 6 7

Demands that subordinates do high quality work 1 2 3 4 5 6 7

Is successful in getting people to work together 1 2 3 4 5 6 7

Supports me in getting my job done 1 2 3 4 5 6 7

Can be relied upon to do what he/she says he/she will do 1 2 3 4 5

Often lets me know how well I am performing my job 1 2 3 4 5 6 7

Encourages people to participate in important decisions 1 2 3 4 5 6 7

Encourages people to speak up when they disagree with a decision 1 2 3 4 5 6 7

Allows people to use their own judgment in solving problems 1 2 3 4 5 6 7

Is making full use of my work knowledge and capabilities 1 2 3 4 5 6 7

Helps subordinates with their personal problems 1 2 3 4 5 6 7

Is concerned about me as a person 2 3 4 5 6 7

Feels that each subordinate is important as an individual 1 2 3 4 5 6 7

295

SECTION 10

The quality of my work is important to the success of the organization

There are strong incentives for me to improve the quality of my work

I have specific ideas about how to improve the quality of work in my group

The quality of my work affects the work of other people in

If I exerted more effort I could improve the quality of my work

Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

SECTION 11 (Please circle the number that best reflects what you think)

How much influence do you have over day to day work decisions that affect you? A great deal 5 Quite a lot 4 Some 3 A little 2 None 1

How much influence would you like to have over day to day work decisions that affect you?

A great deal 5 Quite a lot 4 Some 3 A little 2 None 1

296


Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

This division of is committed to quality

Other departments can be relied upon to do as they say they will do

Other departments put the overall organization's benefit before their own benefit

Overall, there is a lot of cooperation between departments

Continuous improvement is essential for the future success of this site

I feel I am really part of my workgroup

The people in my workgroup encourage each other to work as a team

There are feelings among members of my workgroup which tend to pull the group apart

Do you think that further quality training would help you in your work? Yes No 0 Not sure 9

SECTION 13

Are you presently a member of an operational Continuous Improvement Group within this site? Yes 1 No 0

IF YOUR ANSWER IS YES PLEASE GO TO PART A

Have you ever participated in a Continuous Improvement Group within this site? Yes 1 No 0

Would you like the opportunity to participate in a Continuous Improvement Group within this site?

Yes 1 No 0

IF YOUR ANSWER IS YES PLEASE GO TO PART B

IF YOU ARE NOT ANSWERING PARTS A OR B, PLEASE GO TO THE END 297

PART A

Is it helping you in any way in your work?

To a very great extent 5 To a great extent 4 To some extent 3 Not much 2 Not at all 1

In general, how satisfied are you with the support your group gets for improvements?

Very satisfied 5 Satisfied 4 Neither satisfied nor dissatisfied 3 Dissatisfied 2 Very dissatisfied 1

Do you think training would help your group make improvements?


In general, how satisfied are you with the recognition your group gets for improvements


To what extent does participation in Continuous Improvement Group make use of your abilities and work knowledge?


298

PART B

Do you think it would help you in any way in your work?


To what extent do you think participation in a Continuous Improvement Group would make use of your abilities and work knowledge?


Do you think participation in a Continuous Improvement Group would help make improvements in your work area?


Thank you very much for your cooperation

299

SUPERVISOR 1ST ROUND QUESTIONNAIRE


THE INFORMATION CONTAINED IN THIS QUESTIONNAIRE SHALL REMAIN COMPLETELY CONFIDENTIAL. No ONE WITHIN WILL SEE ANY OF YOUR RESPONSES.

SECTION 1 What is your job title? (Please circle one number)

Module Leader 20 Cell Leader 23 Supervisor 21 Manager 24 Section Leader 22 Other



Within your Module/Department, what cell or section are you in?



What year did you join


How old are you?


300



Very dissatisfied


Not sure Moderately satisfied

Very satisfied

Extremely satisfied

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7



Your fellow workers


Your immediate boss


Your rate of pay



Industrial relations between management and workers




Your hours of work


Your job security


301

SECTION 3 PLEASE CIRCLE THE NUMBER THAT BEST SHOWS HOW YOU FEEL Regarding the site:

Strongly disagree

Disagree quite a lot



Agree quite a lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7




To know my own work had made a contributiion to the good of would please me







Not sure about its

importance


Fairly important

Very important

Extremely important

maximum 1 2 3 4 5 6 7

Achieving something you personally value 1 2 3 4 5 6 7

The opportunity to make your own decisions 1 2 3 4 5 6 7

Challenging work 1 2 3 4 5 6 7

Extending your range of abilities 1 2 3 4 5 6 7

302


Strongly disagree

Disagree quite a

lot



Agree quite a lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7










Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7







303

SECTION 7 PLEASE INDICATE WHAT YOU THINK ABOUT THE FOLLOWING STATEMENTS BY CIRCLING THE MOST APPROPRIATE NUMBER

Regarding management in this site:

Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

5 1 2 3 4 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7












304



Very dissatisfied

Dissatisfied Rather Dissatisfied

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

2 3

1 2 3

Neither Fairly Satisfied Very satisfied nor satisfied satisfied dissatisfied

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

as an employer


Your present job






SECTION 9 Please describe the person you usually report to by circling the number that best reflects what you think: The person I normally report to:

Strongly disagree



Slightly agree


1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7






Gives priority to finishing work on time rather than the quality of work

Sets example of quality performance in his/her day to day activities

305

Demands that people give their best

Strongly disagree



Slightly agree


effort 1 2 3 4 5 6 7






Can be relied upon to do what he/she says he/she will do 1 2 3 4 5 6







Is concerned about me as a person 1 2 3 4 5 6 7


306

SECTION 10






Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7


How much influence do you have over day to day work decisions that affect you? A great deal 5 Quite a lot 4 Some 3 A little 2 None 1



307


Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7






Most of my colleagues make continuous improvement a top priority in their work

My colleagues are receptive to improvement ideas from other work areas

Do you think that further quality training would help you in your work?

Yes 1 No 0 Not sure 9

SECTION 13

Are you presently a member of an operational Continuous Improvement Group within this division (comprising people from different sections who are not part of your normal work group?)

Yes 1 No 0

IF YOUR ANSWER IS YES PLEASE GO TO PART A

Would you like the opportunity to participate in a Continuous Improvement Group within this division (comprising people from different sections who would not be part of your normal work group?

Yes 1 No 0

IF YOUR ANSWER IS YES PLEASE GO TO PART B

IF YOUR ANSWER IS 'NO' TO THE TWO QUESTIONS PLEASE GO TO SECTION 14

308

PART A

Is it helping you in any way in your work?


In general, how satisfied are you with the support your group gets for improvements?


Do you think training would help your group make improvements?


How satisfied are you with the quality improvements being implemented?


PART B

Do you think if would help you in any way in your work?


To what extent do you think participation in a Continuous Improvement Group with people from different sections contribute to quality improvements?


309

SECTION 14

The following answers are general statements, they do not contain right or wrong answers. They are different points of view.

Please indicate what you think about the following statements:


disagree

The best way for a boss to get things done is to use personal authority to direct people

The average person prefers to be directed

Most employees in any organizations do not possess the potential to be "self starters" on the job

The average person wishes to avoid responsibility

Giving greater independence to most employees would be bad or the organization

Enven increased pay is usually not enough to overcome people's inherent dislike of work

People are primarily self motivated and self controlled

The average person can find work a source of satisfaction

Most employees are capable of exercising a certain amount of autonomy and independence on the job

The potential of the average person is much greater than typically recognised by organizations today

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

6 7

6 7

6 7

6 7

6 7

6 7

6 7

6 7

6 7

6 7

Thank you very much for your cooperation

310

EMPLOYEE 2ND ROUND QUESTIONNAIRE



SECTION 1

What is your job title? (Please circle one number)

Module Operator 1 Purchase Progress Controller 5 Module Craftsman 2 Analyst 6 Materials Controller 3 Buyer 7 Engineer 4 Clerical 8

Other


Finance & Planning 1 Alternators 8 Engineering 2 FVE/CSG 9 Quality 3 CA45 10 Market Development 4 Other Starters 11 Customer Interface 5 Process 12 Service 6 Thermostart 13 Inline 7 • Supplies 14

Within your Module/Department area, what cell or section are you in?


How long have you been in present job?



How old are you?


311



Very dissatisfied


Not sure

Moderately satisfied

Very satisfied

Extremely satisfied

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7



Your fellow workers


Your immediate boss


Your rate of pay



Industrial relations between management and workers in your firm




Your hours of work


Your job security


312

SECTION 3 PLEASE CIRCLE THE NUMBER THAT BEST SHOW HOW YOU FEEL

Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7











Not sure about its

importance


Fairly important

Very important

Extremely important

maximum 1 2 3 4 5 6 7





313


Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7










Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7







314


Regarding management in this division:

Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7












315

Overall, how satisfied are you with: Very Dissatisfied Rather

dissatisfied Dissatisfied

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

as an employer


Your present job







Neither satisfied nor dissatisfied

Fairly satisfied

Satisfied Very satisfied

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7



Strongly disagree



Slightly agree


1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7






316

Gives priority to finishing work on

Strongly disagree



Slightly agree


time rather than the quality of work 1 2 3 4 5 6 7

Sets example of quality performance in his/her day to day activities 1 2 3 4 5 6 7







Can be relied upon to do what he/she says he/she will do 1 2 3 4 5 6 7









317

SECTION 10






Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5

1 2 3 4 5 6 7


How much influence do you have over day to day work decisions that affect you?




318


Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7






I feel I am really part of my workgroup

The people in my workgroup encourage each other to work as a team

There are feelings among members of my workgroup which tend to pull the group apart

I am willing to put myself out to help my workgroup



319

SECTION 13 Please indicate what you think by circling the most appropriate number

COMPARED TO A YEAR AGO

My performance on the job has improved

Communication between management and employees has improved

Top management is more committed to Total Quality

Improvements have been made in how I do my job

Visible progress has bee made in improving things at this site

Top management is more supportive of suggestions to improve the way things are done around here

The performance of my work area/department has improved

The level of cooperation between management and employees has improved

Total Quality is a greater priority at this site

People are encourage more to say how they think things could be done better

There is greater contact between management and employees

Strongly disagree



Slightly agree


1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

SECTION 14 Please circle the number that best reflects what you think.


disagree

I would contribute to improving things around here regardless of total Quality

The welfare of employees is taken very seriously at this site

My knowledge of Total Quality has increased my efforts to find ways of improving things around here

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

320

Strongly Disagree disagree

I often put forward ideas and suggestions without expecting extra reward

1

2

How would you have answered this question a year ago? I often put forward ideas and suggestions without expecting extra reward 1 2

In my work area I am always looking for ways to prevent mistakes 1

2

How would you have answered this question a year ago? In my work area I am always looking for way to prevent mistakes 1 2

I am not paid to think of ways of improving things around here 1 2

How would you have answered this question a year ago? I am not paid to think of ways of improving things around here 1 2

Looking for ways of improving how things are done around here is part of my job

1

2

How would you have answered this questions a year ago? Looking for ways of improving how things are done around here is part of my job 1 2

I have put a lot of effort into thinking about how I can improve my work

1

How would you have answered this question a year ago? I have put a lot of effort into thinking about how I can improve my work 1 2

I am strongly committed to Total Quality 1

2

How would you have answered this questions a year ago? I am strongly committed to Total Quality 1 2

SECTION 15 This section asks your view on the following statements at the present time and also what you think your view would have been a year ago. {Please circle the number that best reflects what you think)

Slightly disagree

3


4

Slightly agree

5

Agree

6

Strongly agree

7

3 4 6 7

3 4 5 7

3 4 5 6 7

3 4 5 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

321

How you have answered this question a year ago? To know that I had made a contribution to improving things around here would please me

2 3 4 5 6 7

2 3 4 5 6 7

To know that I had made a contribution to improving things around here would please me


disagree

This section asks you how important you think the following features would be in an ideal Total Quality organization and how important they are in practice in this site.

Please rank all features in order of importance from 1 to 5. 1 being the most important, 2 being the second most important , and 5 being the least important.

AN IDEAL TOTAL QUALITY ORGANIZATION

SITE

Cost reduction

Product quality

Morale

Customer satisfaction

Participation

322


COMPARED TO A YEAR AGO, In general, people at this site:

Strongly Disagree Slightly Neither disagree disagree agree nor

disagree

Are more willing to put forward ideas and suggestions without expecting extra reward 1 2 3 4

Exert greater effort in looking for way to prevent mistakes 1 2 3 4

Put greater effort into thinking about how they can improve their work 1 2 3 4

Take greater pride in knowing they had made a contribution toward improving things around here 1 2 3 4

Are more quality conscious 1 2 3 4

SECTION 17

(Please circle the appropriate number)

Do you have the same immediate boss as you did when you completed the previous questionnaire? Yes 1

Has the content of your job changed substantially since you completed the previous questionnaire? Yes 1

Have you changed jobs since you completed the previous questionnaire? Yes 1

Slightly agree


5 6 7

5 6 7

5 6 7

5 6 7

5 6 7

No 0

No 0

No 0

323

SECTION 18 Here are some views on Working Together To Win (WTTW).

Strongly disagree



Slightly agree


WTTW is a top priority at this site 1 2 3 4 5 6 7

There is a lot of active support for WTTW among managers 1 2 3 4 5 6 7

There is a lot of active support for WTTW among workers in general 1 2 3 4 5 6 7

My immediate boss is strongly committed to WTTW 1 2 3 4 5 6 7

My immediate boss involves me in WTTW 1 2 3 4 5 6 7

My immediate boss has changed his/her behaviour as a result of WTTW 1 2 3 4 5 6 7

Management and employees will benefit equally from WTTW 1 2 3 4 5 6 7

WTTW will benefit me in my job 1 2 3 4 5 6 7

WTTW has improved communications between management and employees 1 2 3 4 5 6 7

WTTW has resulted in better relations between management and employees 1 2 3 4 5 6 7

WTTW has resulted in greater teamwork between management and employees 1 2 3 4 5 6 7

WTTW is an appropriate way to bring about the type of change needed at this site 1 2 3 4 5 6 7

WTTW is no better or wore than previous initiatives 1 2 3 4 5 6 7

WTTW is not part of my job 1 2 3 4 5 6 7

There is no benefit for me in WTTW 1 2 3 4 5 6 7

WTTW is a management initiative to get people to do more work 1 2 3 4 5 6 7

324

How well informed do you feel about the aims and objectives of Working Together To Win?

Very well informed 5 Fairly well informed 4 Reasonably well informed 3 Not very well informed 2 Not at all informed 1

How well informed do you feel about what is expected of you in Working Together To Win?



To what extent are you presently participating in Working Together To Win (WTTW) activities?


AT THE PRESENT TIME:


disagree

I am committed to participating in WTTW activities

WTTW allows me to contribute to improvements

WTTW has no effect on my job performance

1 2 3 4 5

1 3 4 5

1 2 3 4 5

WTTW has no effect on the performance of my work area/department 1 2 3 4 5

WTTW gives me greater influence in what goes on in my work area/department 1 2 3 4 5

6 7

6 7

6 7

6 7

6 7

Would you like the opportunity for greater participation in WTTW activities?

Yes 1 No 0 IF YOUR ANSWER IS YES PLEASE ANSWER THE FOLLOWING QUESTIONS. IF YOUR ANSWER IS NO PLEASE GO TO SECTION 20.

325

Strongly disagree



Slightly agree


1 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

WTTW will improve my job performance

WTTW will allow me to contribute to improvements

WTTW will improve the performance of my work area/department

SECTION 20 This section is about Total Productive Maintenance (TPM). If this does not apply to you (i.e. you work in a non-manufacturing area) please go to THE END.

Have you received training in Total Productive Maintenance (TPM)? Yes 1 No 0

Are you implementing TPM practices in your job?

Yes 1 No 0

IF YOU HAVE ANSWERED YES TO THE LAST QUESTION PLEASE ANSWER THE FOLLOWING QUESTIONS. IF YOUR ANSWER IS NO PLEASE GO TO SECTION 20b ON THE FOLLOWING PAGE.

TPM gives me greater responsibility

Strongly disagree



Slightly agree


in my job 1 2 3 4 5 6 7

TPM gives me greater control over the quality of my work 1 2 3 4 5 6 7

TPM gives me more variety in my job 1 2 3 4 5 6 7

TPM give me greater involvement in my work 1 2 3 4 5 6 7

TPM allows me greater influence in what goes on in my work area 1 2 3 4 5 6 7

My immediate boss in genuinely committed to TPM 1 2 3 4 5 6 7

There is strong management support for TPM 1 2 3 4 5 6 7

TPM has no effect on my job performance 1 2 3 4 5 6 7

TPM benefits me in my work 1 2 3 4 5 6 7 PLEASE GO TO THE END

326

SECTION 20b

Would you like the opportunity to implement TPM in your job?

Yes 1 No 0

IF THE ANSWER IS YES PLEASE ANSWER THE FOLLOWING QUESTIONS:

TPM will give me greater

Strongly disagree



Slightly agree


responsibility in my job 1 2 3 4 5 6 7

TPM will give me greater control over the quality of my work 1 2 3 4 5 6 7

TPM will give me more variety in my job 1 2 3 4 5 6 7

TPM will give me greater involvement in my work 1 2 3 4 5 6 7

TPM will allow me greater influence in what goes on in my work area 1 2 3 4 5 6 7

TPM will have no effect on my job performance 1 2 3 4 5 6 7

TPM will benefit me in my work 1 2 3 4 5 6 7

THANK YOU VERY MUCH FOR YOUR COOPERATION

327

SUPERVISOR 2ND ROUND QUESTIONNAIRE



SECTION 1

What is your job title? (Please circle one number)

Module Leader 20 Cell Leader 23 Supervisor 21 Manager 24 Section Leader 22 Other



Within your Module/Department area, what cell or section are you in?

Not appropriate 0

Cell/Section Name/No




How old are you?


328



Very dissatisfied


Not sure

Moderately satisfied

Very satisfied

Extremely satisfied

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7



Your fellow workers


Your immediate boss


Your rate of pay



Industrial relations between management and workers in your firm




Your hours of work


Your job security


329

SECTION 3 PLEASE CIRCLE THE NUMBER THAT BEST SHOW HOW YOU FEEL Regarding the site:










Not particul arly

important

Not sure about its

importance


Fairly important

Very important

Extremely important

maximum 1 2 3 4 5 6 7





Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

330

SECTION 5 PLEASE CIRCLE TRE NUMBER THAT BEST SHOWS HOW YOU THINK

Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

If I got into difficulties at work I know my colleagues would try and help out


I have full confidence in the skills of my colleagues

Most of my colleagues can be relied upon to do as they say they will do

I can rely on other colleagues not to make my job more difficult by careless work

Most of my fellow colleagues would get on with the job even if supervisors are not around

My colleagues are genuinely committed to improving quality

My colleagues have specific ideas about how to improve the quality of their work


Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7







331


Regarding management at this site:

Strongly disagree

Disagree quite a

lot



5

Agree quite a

lot

Strongly agree

1 2 3 4 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7












332



Very Dissatisfied Rather Neither Fairly Satisfied Very dissatisfied Dissatisfied satisfied nor satisfied satisfied

dissatisfied

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

4 5 6 7

as an employer


Your present job









disagree






1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

333

Gives priority to finishing work on

Strongly disagree



Slightly agree


time rather than the quality of work 1 2 3 4 5 6 7

Sets example of quality performance in his/her day to day activities 1 2 3 4 5 6 7




Demands subordinates do high quality work 1 2 3 4 5 6 7



Can be relied upon to do what he/she says he/she will do 1 2 3 4 5 6 7




Allows people to use their own judgement in solving problems 1 2 3 4 5 6 7





334

SECTION 10






Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7


How much influence do you have over day to day work decisions that affect you?



A great deal 5 Quite a lot 4 Some 3 A little 2 None

335


Strongly disagree

Disagree quite a

lot



Agree quite a

lot

Strongly agree

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7






Most of my colleagues make continuous improvement a top priority in their work area

My colleagues are receptive to improvement ideas from other work areas



336

SECTION 13 The following answers are general statements, they do not contain right or wrong answers. They are different points of view.

Please indicate what you think about the following statements:

Strongly disagree



Slightly agree


1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 '7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

The best way for a boss to get things done is to use personal authority to direct people

The average person prefers to be directed

Most employees in any organizations do not possess the potential to be "self-starters" on the job

The average person wishes to avoid responsibility

Giving greater independence to most employees would be bad for the organization

Even increased pay is usually not enough to overcome people's inherent dislike of work

People are primarily self motivated and self controlled

The average person can find work a source of satisfaction

Most employees are capable of exercising a certain amount of autonomy and independence on the job

The potential of the average person is much greater than typically recognized by organizations today

337

SECTION 14 Please indicate what you think by circling the most appropriate number

COMPARED TO A YEAR AGO

Strongly agree



Slightly agree


1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

My performance on the job has improved

Communications between management and employees has improved

Top management is more committed to Total Quality

Improvements have been made in how Ido my job

Visible progress has been made in improving things at this site

Top management is more supportive of suggestions to improve the way things are done around here

The performance of my work area/department has improved

The level of cooperation between management and employees has improved

Total Quality is a greater priority at this site

People are encouraged more to say how they think things could be done better

There is greater contact between management and employees



disagree

I would contribute to improving things around here regardless of Total Quality

The welfare of employees is taken very seriously at this site

My knowledge of Total Quality has increased my efforts to find ways of improving things around here

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

338

Strongly Disagree disagree

I often put forward ideas and suggestions without expecting extra reward

1

2

How would you have answered this question a year ago? I often put forward ideas and suggestions without expecting extra reward 1 2

In my work area I am always looking for ways to prevent mistakes 1

2

How would you have answered this question a year ago? In my work area I am always looking for way to prevent mistakes 1 2

I am not paid to think of ways of improving things around here 1

2

How would you have answered this question a year ago? I am not paid to think of ways of improving things around here 1 2

Looking for ways of improving how things are done around here is part of my job

1

2

How would you have answered this questions a year ago? Looking for ways of improving how things are done around here is part of my job 1 2

I have put a lot of effort into thinking about how I can improve my work

1

2

How would you have answered this question a year ago? I have put a lot of effort into thinking about how I can improve my work 1 2

I am strongly committed to Total Quality 2

How would you have answered this questions a year ago? I am strongly committed to Total Quality 1 2

SECTION 16 This section asks your view on the following statements at the present time and also what you think your view would have been a year ago. (Please circle the number that best reflects what you think)

Slightly disagree

3


4

Slightly agree

5

Agree

6

Strongly agree

7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

3 4 5 6 7

339

How you have answered this question a year ago? To know that I had made a contribution to improving things around here would please me

2 3 4 5 6 7

2 3 4 5 6 7

To know that I had made a contribution to improving things around here would please me


disagree

This section asks you how important you think the following features would be in an ideal Total Quality organization and how important they are in practice in this site.

Please rank (from 1 to 5, 1 being the most important)

AN IDEAL TOTAL QUALITY ORGANIZATION

SITE

Cost reduction

Product quality

Morale

Customer satisfaction

Participation

340


COMPARED TO A YEAR AGO, In general, people at this site:

Strongly Disagree Slightly Neither disagree disagree agree nor

disagree

Are more willing to put forward ideas and suggestions without expecting extra reward 1 2 3 4

Exert greater effort in looking for way to prevent mistakes 1 2 3 4

Put greater effort into thinking about how they can improve their work 1 2 3 4

Take greater pride in knowing they had made a contribution toward improving things around here 1 2 3 4

Are more quality conscious 1 2 3 4

SECTION 18 (Please circle the appropriate number)

Do you have the same immediate boss as you did when you completed the previous questionnaire? Yes 1

Has the content of your job changed substantially since you completed the previous questionnaire? Yes 1

Have you changed jobs since you completed the previous questionnaire? Yes 1

Slightly agree


5 6 7

5 6 7

5 6 7

5 6 7

5 6 7

No 0

No 0

No 0

SECTION 19 Here are some views on Working Together To Win (WTTW). Please circle the number that best reflects what you think.

Strongly disagree



Slightly agree


WTTW is a top priority at this site 1 2 3 4 5 6 7

There is a lot of active support for WTTW among managers 1 2 3 4 5 6 7

There is a lot of active support for WTTW among worker in general 1 2 3 4 5 6 7

My immediate boss is strongly committed to WTTW 1 2 3 4 5 6 7

My immediate boss involves me in WTTW 1 2 3 4 5 6 7

341

My immediate boss has changed his/her behaviour as a result of

Strongly disagree



Slightly agree


WTTW 1 2 3 4 5 6 7

Management and employees will benefit equally from WTTW 1 2 3 4 5 6 7

WTTW will benefit me in my job 1 2 3 4 5 6 7

WTTW has improved communications between management and employees 1 2 3 4 5 6 7

WTTW has resulted in better relations between management and employees 1 2 3 4 5 6 7

WTTW is an appropriate way to bring about the type of change needed at this site 1 2 3 4 5 6 7

WTTW is not better or worse than previous initiatives 1 2 3 4 5 6 7

WTTW is not part of my job 1 2 3 4 5 6 7

There is no benefit for me in WTTW 1 2 3 4 5 6 7

WTTW is a management initiative to get people to do more work 1 2 3 4 5 6 7

How well informed do you feel about the aims and objectives of Working Together to Win?


How well informed do you feel about what is expected of you in Working Together To Win?


342

SECTION 20a

This section is about Total Productive Maintenance (TPM). If TPM does not apply to your work area (i.e. you are in a non-manufacturing work area) please go to SECTION 21.

Are you implementing TPM practices in your work area? Yes 1 No 0

IF YOUR ANSWER IS YES PLEASE ANSWER THE FOLLOWING QUESTIONS.

IF YOUR ANSWER IS NO PLEASE GO TO SECTION 20b.


disagree

TPM has had no effect on the performance of my work area 1 2 3 4 5 6 7

TPM has had no effect on my job performance 1 2 3 4 5 6 7

TPM allows the people who work for me to become more involved in their work 1 2 3 4 5 6 7

TPM benefits me in my work 1 2 3 4 5 6 7

PLEASE GO TO SECTION 21

SECTION 20b

If there is a plan to implement TPM in your work area please answer the following questions. Otherwise please go to Section 21 on the next page.

Strongly disagree



Slightly agree


1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

TPM will have no effect on the performance of my work area

TPM will have no effect on my job performance

TPM will allow the people who work for me to become more involved in their work

TPM will benefit me in my work

343


Strongly disagree



Slightly agree


I liked the WTTW training 1 2 3 4 5 6 7

The WTTW training lived up to my expectations 1 2 3 4 5 6 7

I learned a lot from the WTTW training 1 2 3 4 5 6 7

There is a lot of lip service paid to WTTW 1 2 3 4 5 6 7

I feel I have participated in the design of WTTW 1 2 3 4 5 6 7

I have autonomy with regard to implementing WTTW in my work area 1 2 3 4 5 6 7

My colleagues would say I am committed to WTTW 1 2 3 4 5 6 7

WTTW is an important part of my job 1 2 3 4 5 6 7

It has been quite easy to transfer the WTTW training to the workplace 1 2 3 4 5 6 7

My immediate boss supports me in my attempts to practice the principles of WTTW 1 2 3 4 5 6 7

There is a clear consensus as to what WTTW is about 1 2 3 4 5 6 7

WTTW has had a major impact on how I manage the people who work for me 1 2 3 4 5 6 7

What I have learned form the WTTW training has been very useful in my job 1 2 3 4 5 6 7

I am practising the principles of WTTW 1 2 3 4 5 6 7

WTTW has provided the opportunity for greater contact between different work areas/departments

1 2 3 4 5 6 7

I have kept the people who work for me well informed about WTTW 1 2 3 4 5 6 7

344

Strongly disagree



Slightly agree


1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

WTTW adds too much to my workload

There is a lot of pressure here to practice the principles of WTTW

I am well informed of my role in WTTW

The people who work for me are willing to participate in WTTW

WTTW will definitely achieve its objectives

WTTW has resulted in a greater sharing of information between work areas/departments

WTTW has had no effect on the performance of my work area

I am rewarded for my participation in WTTW

WTTW has increased my knowledge of the needs of other work areas/departments

There are strong incentives for me to practice the principles of WTTW

WTTW has had a major impact on how I deal with other work areas/departments

My efforts to pursue WTTW are recognised

My behaviour has changed as a result of WTTW

I have involved the people who work for me in WTTW

WTTW has had no effect on my job performance

To date WTTW has been successful

THANK YOU VERY MUCH FOR YOUR COOPERATION

345

APPENDIX 2: Sample characteristics for site 2

Site 2

1. Mean Age (S.D) 37.5 years (10.0) 2. Length of time in present job (S.D) 4.2 years (3.5) 3. Length of service with organization (S.D) 7.6 years (6.2) 4. Length of service at site (S.D) 7.4 years (5.77) 5. Gender

Male (% of sample) 207 (91%) Female (% of sample) 21 (9%)

6. Job Categories Manufacturers (% of sample) 113 (49.4%) Engineers & Research (% of sample) 43 (19.0%) Clerical/Administrative (% of sample) 27 (11.7%) Supervisors/Managers (% of sample) 45 (19.9%)

346

APPENDIX 3: Intercorrelations

Table I

Table II

Table III

Table IV

Table V

Table VI

Table VII

Table VIII

Table IX

Table X

Table XI

Table XII

Table XIII

Antecedents of team orientation at time 1 (chapter 4)

Antecedents of team orientation at time 2 (chapter 4)

Evaluation model of team orientation at time 2 (chapter 5)

Evaluation model of team orientation - change data (chapter 5)

Antecedents of commitment to improvement at time 2 (chapter 6)

Antecedents of commitment to improvement - change data

(chapter 6)

Evaluation model of commitment to improvement at time 2

(chapter 7)

Evaluation model of commitment to improvement - change data

(chapter 7)

Participation model (chapter 8)

Assessment model (chapter 9)

Supervisor participative style model (chapter 10)

Supervisory commitment to quality model (chapter 10)

Performance improvement model (chapter 11)

347

* * * * * * VD 00 VD

N on

. . .

* * * * * * * * fq CD CD

M vl • •

4(1■1 CN 00 Vn vn

ON CD ,t

. . . .

* I ,t vn C

O CD c)

*

I * ,0 un 00 N N ,-, CD C) CD CD CD CD . . . . .

* * * *

I CD r- ON s CD. CD CD

•

*

cD

* * * * *

* * * * * * * 0,1 ,. 00 O M CD -. VD "cr

M. un C) C>

* * * *

,t on 00 N N ,r M R R CD CD CD.

; O cr 2. o

c.) t' "

= '" 15,

bk W 0 0 4 m & 5 :,) E1. E

o w cn w E z ., .°

> .- '''.

w tt 0. *= -F-5 71 4.) (do =

,.. e., ,„ 0 5 .- = o w .4 ct Vi i- ..= ;,.. g E. 4. F: .?Jo 6- = e .- 7. ez: E

t: tn E0 . 4 ....

az cu = = = .... 'A 73 tl''. ,9 et 77'. In "ml La CD 0 0 OP W w 0 M 1. 0 0 . CI) V N,/ P, '' 0 .4 Imo on E. CV coo -,k.- E.

C. Cli o6 cr; 7:4 0 4: 4: .4 N M4 tri ■: c-:

CO •:

348

cr■

00

N

‘D

N

* * * * * **y * * 1-1 I VD on 00

MM tin . . . .

* * * * * * * * •7 N 00 kr) •7V

,r-)n. . •. .

I on CA cr, ni ,f I C: CR cR) C' 'R. CR. CR.

I •

* * I VD N 'ch N— on

CR CR (R)

* * * * 14'n N 7 CD CD N 00 N ",1"

cnon C) c) O c) . . . . . . .

* *

M 00 N N CO O 0 0 C> 0 CD 0 C> . . . . .

* * * *

* * * * * * IC- ,-. N 00 VD ,t N C' on N N ,--, •-. '" C) CD 0 CD CD CA,-.. CD

. .

*

* VD 00 N ken on tin N CrNon

NI C) CD CD CD CD CD CD CD

. . . . . . . . . . . .

* * * * * * * * * * * * * * * * * * -,zI. CD 00 VD 00 .1. CA r— cq cr, c) v.-) VD

In c::, ■--, NNc) 0 CD ,-ImN ,-I

* * * * * * in ,. 00 ,.

1 ic!..1 N 4 * *

* *1 ,-I O\\ 0 (0 con 1.0 I 01 C) 01 ,. ,. ^, N C) CD CD 0 CD CD CD

. . . . . 1 . ,

CL

RS (in

On O-1

G2.• ca -X

Job

Tenu

re

0 -. cr a 0 cr, -4.. 0 0 -LI

" a) 4-+ .y. = W' = 5 on 00

ct _,-. 0 0 cn ,,, .0. 4.) O.) w co) .) E t) 0 = 0 ... •.47i

.1 - ") C-14 CI" ❑ 1' = 0 ri, 1...

50 "; 0 0 ,... 1... 0

Q. -0 .2 5 ',..- 4) >) 6- o

.. :4- = 0 0 „t • .,., ts 0 0 -0 VI) E* 0 . -4 --, 0 0. 0 5 5 5 a re) 7,4"-' at : 1-, 5

0.) 0 : _0 47... _, ow al) 0 0.) p • • • ct 6, "' = IL-■ E 0.) -el CD 04 on < a -(1 (i.) E- ■ 0' cn i.,

,4 ri egi 4 ui vi, t-: of3 c; 4 7' --i ej 1 1-4 Ill

349

* * * * * * e-1 co kr) I n1-. trl

* * * O Nt N

4-1 v■I N • • • •

t' • ■ 0 0 CX 0 0 0" "4. 0 ..E

'4-' 0 CL1 0 14 E CA cu cp

CS

.0 = 0 -- = •... > ,. -5 :-. tm = t ° ° ° ,, 0 ,,, R' 2 i c % -,-, l 0 ., .... 0 C. 0 = ..,- a) cu 0

0 ,... .: 4... — ..0 .0 . ..., , , . c, 0 — = 0 © 0: -0 t'L - .--- `t7 -,1; .rt 0° 2

„0 0 0 0 .0 ..4."5 0 CI O. .6:i 0,, c: . b.0 CD 0 , •E E 0 • ..... ,,tt 7 = ,.., cc„

)— -tt (..7 ■4 0-, <1 <1 d cA E.. CY cip ■- E.*

,-4 (..,i toi .. ,ci: tol ■.c:; t..-: 06 a; ,-4° ,—.71 - ,—.1`.6 -7 : 4

350

0 161

00

N

LOD

kr)

eF

N

7-C

* * * * * * * * * * * * l';'1 :0 N 01 \O V1 V1 C V1 V1 CA.. .. CA 01 01 CA

. . . .

* * * * * * * * * * *

00 * * * *

.. V/ V1 .. CA 00 01 N 01 ,t 01 'eh e. .. CD CA CA 01 01 . . . . . .

* * * I N ,t CA M CA .. 41 N CD CD CA

I .. CA CD CD CD CD op • • " • • " " •

* * * * * * * * * * * * * *

M Ch N V1 CD 00 'et N N Q CA Cl Cl • • • • •

* * * *

I .h 00 01 01 00 VD 01 .. 00 00 in VD \C, NCIO 0 0 000 CD 0 0 CD 0

* * * * * * * \ID CA O VD C) OA CA N In CA VD N

0

.. on CD CD CD CD CD CD CD CD . . . . . . . . .

* N on C) VD N on N •. N •• 0 01

CA CD CD CD CD CD CD CD CD CD CD CD CD c>

. . . e e o • e j j

CA C) VD CD. (D.

CA CD h ,t VD h CD CD CD CD op op op 0.

* * * * * * * I VD ,t •. O. ,f CD .. h ,..1D on N 01 CA .. CD N ..

CD ,t CA CD CD CD .. CD CD CD ,-, .. CD O,-I ,-, ,-,

* * * * *

* * * * * * I .1- vA ,, VD 'et CT .71. ,... N .. .. CD 01 el- In .i- VD CA

N 0 0.1 ,-I .. 0, cA CD CA •-o CD CD CD CD CD CD CD CD • • • 1 1 o ■ 1

11 =

1; CS' = o tr, -4-, '-'12 o '5

44 '

rat 0 0 C4 0 E a) a) bk 5

F:1 ag = CS a)

as 0 4.) a% • ..., E 0 1:

L O rnCL cn ..-. E 0

.= 0 —

4. = 0 = ...4 '4... OP = t .2 i...

S. C - a) ). 'ES ... o C Ow 4.1° 0 et -I. 4" GO

,A = 0) as Ow t 0 +4-'

a.) GO > -5 .0 .1:2 as r-b = , 4.) z

;.., s... 0 0) , t g

C.

E 4, crs "0 "0 0 ^.

0 0 0° a) a) .0 .4:1 1.., .— .— .— 11., '260 2 1='. ;-. „t' 5 . *4 tnu 4) ;-' = o 0)

EE" -c, 461 E-1 . N. • .6 6 cl 0 -.= 23 eo ut 4'0 74 a) z t

az = a.) = = an = 1. y.., i.., ., '..4= 0 etWajo C e Z -' "-'04.) a) Ce:le°gw4E 0

)e: ",t (.. 14 ee '<1 <1 <1 Pi P0 16 U0 (40 E. 0 00 *0 E.

el: Ili LA t. 06 o■ :4 ai'l :1 e,i ,:i vi ■46 r-:- 06 07A

351

I 0 * * * * en en

* * * * * * en en C\ r) M "I' . .

* * * *

N I et el- CD

N

* * * * * * * * * en Cr, rn m 00

m M m kr-)

iF * * * * * k * * * I vz) Cr N v:-.) kr) I '—'. N cl

* * * * * * * * * * * * * *

I 00 tr) Q1 ton en .1-

. .

* * * * * * * * * * * * * * * *

ca, 0 co •-∎ '.0 '.0 C.1 M 'Tr "Zr . . . . . . .

V) -4•• 0 VI = Z CU CU 10) g E as

.4-' 8 = 0 . , •2'a 75 . c.? ,,,.; .4.5. E• +rce, '..=

.... 4- .E ci) > • o 0.) o. 6, 0) 4 4E' = a)o cs.

L cf) . > 2 .0 0 b ,.. „ 0 ..,, • 4-I 70 .0 0 CD 0 0 O.) ‘) (n cl aj ct) 4 -1 = 1_, 47, = xl J0 04 > ›. ..,_ ca

E.. .0 to, t., ,#) •-, ••.-, •-■ 8 Q . a) .4F .--, = ... -- ,,, c,., = = ,.o .5 9 5 r. 1... s... 0., a o tu) 4.) a.) 0 0 c.) a) = as d c... 1-4 <1 <1 <I A.. P.4 P.( cc ct)

,1 N Met: V) ,46 ir-: 06 cr■ ,-16 ,—H (Ni•—■ ,—■"-;

352

1.0

CE

C.)

CC 4:1 IDA Z:1

• 44.1

5 Cd

V

E V

O x.61 V

5

..47) C11

-1E1 • ■-1 O

td)

O E CC

O •

••■■ V ;.4 it O V 44

4E1

111

C•r1 •-•4

00

1/44D

•tr

N

* N

* * * * ••,I•

* * * CD

* * * C7N 0 CO 8.

■-•4

* * * * Mv 00 tr) 1•■•• CR '1

* * * * * * s.0 co 1/40

Cl cn ■-■ CD CD

*

sjE 0 VI rn 7t* ••th' 0 0 0 C:D CD. . . .

*

N 0 N \ C). C). Q.

* * * * N 1-1 tr) N ,z1" 0 0

"zr 0 M. . C).

* *

cn 00 cr■ N \D •—• N CD

I C)0 0 Cp. C).O Q . . . . 1

* * * * * * *

ti N oto 00 •cf• I 0 00 • • • • • . .

lc) co N kr) rn •ch N N N

0 0 0 0 0 0 0 CD C). . .

* * * * * * * * * * •,:1- 0 00 t:3 00 rn •ct• C:f■

kr) 0 N N O 0 0 0

* * * * * * * * * *

NN•ct kr)(nCr■ \ O d'en 0 M N 0 0 CD Q • • • • • • • •

*4 CO fal

C.0 0 6 0 ca 0 a) 0 a) 0 -4-> 0 C) d) 6 0+ = Cr Cr .,

4. 0 -4. • . 0 C/ > 6 = 0 0.' t °l+) 0 •••■ 0 C Y. 0 4,) ..,t c.) 0 cn E -I-, , 4-. .0 a) 42 _ CU Cr) > '' 4.7 0 >•-, .; (a 0 a.) 0 p

1... 01 0 a a.) ,,-, L., 0 4 w '-' ;-. 0 ',P e..9 L' 0 .4 ..9 ,4 74 = 5 cu 5 E c L. -,.. ,--, .0 s:2 4. - = ow ct 4.. 4-, c..., ..., 4) 4) ''''' 0 0 0 WI 01) 6 = 0 t) = 461 V, • ,= • .-, WI CI 4. Cr , A E E 2 E c E „, 4.) = = = 5 c . -4 a' 0 .. , E z

4 "t" CS ) 1.) ot© <l. <I.c <1.5 <3".' 5t.) -<I.5 . 5

<I <7

c

tva <7 C)—Ei 8a

cS ,--1 ei e*; Ti tri ,'I e4 rei 4 Lri t--: oe, cn ,--1 ,-1 ,-4 ,-1 ,—I .,—I

v-i On c:7( 4) V V -4—. .. Ro t:14 cn 4: a<

353

1.

Job Te

nure

O

Cn

CCt 00

CCS

N

O CV

p.■(

rr:t

E VD

0 CI • •—••■ O 40

'4.) :8?

**4

** * * * * * * * * * * * * CD WI CT CD M c4 Ul Cl N cl

* * * * * * * * * * * * * * * 00 CD N 00 00 cn e4 CI el el CI Cl m cl

* * * * * * * * * * * * * * * * * * N N 00 on co VD 'TS: cn en st st ,r Cl .

* * WI sr Or • • •

CD Cl VD CN CD st CD CD CD CD CD CD CD

* * *

* * * * * * VD ,.. VD CN m N ,-. CN N ,-. ,-.

st ,. ,-- CD CD CD CD CD ,. c) O • • • • • • • . • •

* * * *

4-1 N d' 00 W. Cl CD 41 N

mm co co co co co co ..

• • • • •

* * * * * * * * * * * *

en en 00 CT CT ,..-, 00 N ..c. en N N ,--,

. . . . . . • • • • •

* * * * * *

00 VD VD * ,-. * * * * * *

N N en on ..,1- as ,.. N cq N -. CD CD CD CD N CD ,. CD CD N ,--. .

*

00 N 4,

m WI 7r on 0 N 01 CD N

CD CD 0 CD CD CD CD CD CD CD CD

* * * * * * * * * * * * * * * * * * * * * * * * * *

00 VD 00 st st CT VD on

on oo co-, cl co, ten 0 ,-. Nc‘t ,-, co ....on•- ■ cleco co c.) CI

. . . . . . . . . . . . . . . .

* * * * * * * * * * * * * * N N 7r WI (41 01 0 4, C., ,I. (.0 ,t c) 47 Cl ) f(1 ,-, ,--, ,4 N 0 •0 0 0 0 ,--. 0 .

yl VD .4-1 C).

4-.

74 = 0.• 0 0 a... Cr = 4, 4.

t' cr o ..,.. c.4_,

0 =

0

0 E `1) ,tA 0 =

4. 4. :40 2 E -t..-. 74 E ;.., 0 , 4) 0 ., m . 0 •... „.,

o .. ,...,c-, 0 4. • -. m ,

.02 0= E : aa 5 5,

0)

•

.4 4E' .1) pe 0.) ct • 0 0 0 • ,.. c) el.) 4) L., -t0 , s R. c.,.. „, t E E 2 44 441 E 0 e.) ci) à c4 0 . .

L- ., r:L. x) .2 7,1' '' .. . — tq r.4. , > ›. et •• a • r,00060 c)t t ..4 0 ..... -0 pp tel •= •= cu ;_, • .. z 0, i) "'•'

•

C0 CI as 09 FL) = 0 0 0 C O. I:: , 01y to M f:1( 04

'''t 0 1-4 -.) <I <1 -1 0 0-I = 44

.54 z o .' (2, 4 4 _. , t

g v v , 0 ci,.,

0 ks t4 06 0; ,..+ (-4 ri ei ■.16 -: GO -X *

354

v-4

* * I * *

4-1 00 et cD

• •

* * * * * * * * on oo r- r-

vn 71.

* * * * * * * * * In 4-1 7P rn • • • • •

* * * * * * * * * * *

I 00 CI 71' CNI V:) t,

l 0: ." ''' 0 A-. 0 • ,-, c .1 n a ‘) E Cr Wtl? 0 0 O 0 E g ::.' 09 4-' 3.1.

..,..,

0

= = et 0 E 0 E 0 E0.,,=00"tzna. c.)

-. ,.. ,.. e , -,..)0-Ac,.,•-- L

•

0 0 e *, ..4v, y,0 ,..., +., 4., W W C/D •; 0 4.) ° 0 C >, © cu 0 g F, ,., ,.. p, T., 4' W 3. = 1Z O E 0 O

,... O w 0 - c.) 1c%'8 0 2 2 • ,, -g -g .- 0

; › -. E. .. . . g .-- .6. E. .E LI t; g _g — E: E: az ton 0 21 0 0 0 0 04 4. E 0., 01) -6,0 0 04 Ow O tb W 0 • ..-. ..-, W -i- .., = L., = 0 0

<10.4zulo CY 4" 4: ,-4 (.4 M 4 uri 00 cr■ 4 ',-.;' f . ,' '4; .1', :4 4 !.:, 4

; = -01 • 0.

W 0 4

CS 0 ..,

•

1. 1. ° 4. , .=

0 4 .... a)

355

I gl ** * * * * * *

00 h N

CTCD CI .. CD ,, . .

* * * * *

* * *.-.

* * CD Cm kr.

CD .. Cl ,1 CD CD Cl CA

* * * * * *

* * * * * *

CD ,1- .. on vD .. .. rq ..

WI CD CI C4 .. CD CD km on

. . .

00 * * 0 cs 00 I

,. ,, ,t

-, 0 0 .--i r- ,,, ,r, c, ,f

. c> 0 c) 0 0 c). s.„ ct

cl * cet *

1:1 ,r cll I VD CD .. 4n ..

CD CD CD CD CD CD CD CD CD CD

5 0 t,$)

* *

* * * 0, on CD CD CD CD

. . . . . . CM .. .. knn Inn Cl C> CD CD CD CD CD 4i:11 En9

40 ,-, ■

P. P. C.) 1.). ■•,_ t/JD

* *

...P. 4n 1 m on 00 .. N CD N *

I CD .. CD CD CD CD CD ,. .. N VD N ... CD CD CD CD ,. O

* r- .. PM

0). ;)

1 . . . . . . ,

-, * * * * * * * . • ,—• 0

.,,r 1 s ..N 00NNOCM

CD CD CD 0 .. ...1- CM 4n CD Cl N CD CD CD CD CI .. 46

c ,-,

.. .

. O.) lil • CCS

CI) .. .CD w0 * 13

00 1g en 1 in cl v--, en cp cq ,i- N

1 CD CD CD CD CD CD CD CD CD CM -1- .. NON CD CD C) CD CD CD

* pi 75 1 . . . . . .

E 151 * *

* * * * * * * *

* * * * * * * 0 0 1 '71' CD 00 10 00 7). 00 C11 C11 01 C.) Co) CI I .. 14.1 CD ., NN CD. CD CD .. . . . . . . . 8 crS. .T) S. gi NI

:k0 "C , . I .

0 0 0 CI * * * * * * • *

E .•,. r.

1.4 1 C11 C11 `71' 11r) en CM CM 00 ND -. CI *

M CD on ,. ,. .. N CD CD CD . . .

* * * * * * * *

,-. •-. CD 0 CD 0

Ln

v... CD M k•C) IS) In

4.1 . .

7:141) CD c•-•> 4, 4, —

= cut

4 0 , Cr(7.1 0.. ........, it a" ..= 0., o

.- az 1. 4.4 CI 0

0 --, g 1—' .4 mn 0 .4.4

L. 0 5 cu ..E. 0.9 e..) cm 4. = s E

.... -44 g i 1= z +q I, 0 4.4

ca . Z 4a, ,, g —. ..

CI CI 0 =

-4.+ > te cz ..■ .... . Z , c.., CI ■-, a) cr ,..,

a) c -1-' — 4. = )... 6

u -,-■ 4) g ' = ta = = P

s, 0 4. 4.4 ,4, 0 0 0 = •MK Ei .6 N

0 V) -1., 0

V) , E â p. .7., tz > • MI , 4,1 C, 1.)

4. 6 c.,I t P. .E 4) = nu -, E E o a) cP u) = = — a) ci.)

=1. = am a au CI. 0.1) tn 0 .. 0. 0.

f' .0 ' ,,,, -.Elk ) ,n, •,-, •1) ... -5. E t'', ,..", 0 = Cr E E , .- = , ,

-, 0 = ;..cl •_ _

g',, = = .0 5 .E .E .° .E .E .= 1i E : E fa a' _ 0 nj c, • 5 E < C.7 .4 < « « < < 0. < < « *0 *0

N M 4 IA v5 h 06 .; i4 7.:( efie4 4 tri .c.) C. of3

1■1 ,I ,I i• 1■1 ,..1 1•1

ti

O

CT

1. J

ob

Tenu

re

356

I 0 0 i 0 0

* * * 00 0 on c)

* * * * * * * * (-1 CT C) V)

on Cl • • • •

* * * *

vn V) en VD N 0 0

* * * *

* * * * * CT CT V) CT CN c• cl on C).ct on

. . . . .

k P .. . at = = 4. . crt cr. ..= 04 -4.a .= 01 0 -4... = -F. .4.

C '''''' -4... g -5.0 0 -44 cr 4.4 CU = = = E c 0 E a.] •P,1:: :4c 5E,;:-, 73E

a. .,_. E E 8 . 2 t.. E •-

-4... ti) 0 , E 2 4, -51 E

13 0 > w o c =r o -4... 4_4 ...1 al 0.) W C C- a+ W = i. i. C 0 C.4 CU 0 = ›-4 0 0) 0 = =

ti.), . 1,, .,., 0 ,k) . 2 0.4 : : co .: : : a) ...

" 0 61 2 6 .. 0 en z ..... -. o 4, o --4 CP 0 uo > = m > w) 0.0 .-" ." 1. . 0 W i. ;.., 0 4, CD 0 0 gll

0

0 a) c&

c" . t) 6% a = 0.1 ca.: , 4., 5 a) O. t).0 tto g ›, > = "O t a 5 2 = - -- ot ..... _ . . E.I Pc la M. • ■■., •■,, te •... •...5 4.1 :t; C..1 0 .= so y: y.. 4 c:).41 7-1©

0 .1:1 4) = = ,,c) C C C = C C = ct C 0 C = Ow al, ok a) „ 0 .— , .. ". ... .— E E

<4 C.,.. 14 ■-t <1 <3 <1 <1 <1 <1 <1 cr<1 -4 <1 <I 1-c 1-1 -45 V. V. 0 0., 0.o E-( -x 4:

e4 N vi et: 4,3 v:i N06 0 ,': ;=.4 'T( .:11:, 4 ',.' 4 co ac

357

** * * * * * * * * CD Vl CN CD en Cl Cl ol et et

* * * * * * * * * * 00 CD N 00 . 00 0,1 • rl el ri rl

* * * * * * * * * * * * N Vn C4 cl

Me. VD CD ci cl cl

* * * * * * * * * * * * Cl N 00 ON CD et rl rl ri

1. Jo

b Te

nure

* * * * * * * * * * * * * * * * * * * * * * * * * * N CN N 'CC 00 en c) C2, N et et 4-) cn on et eF 41 et on en on C,1 CV

' ' • • • • " • " •

* * I CD Cl VD CN CD 7C 'CC ,71• CN en I CD .. CD CD CD CD CD CD CD CD CD CV C) CD

* * * * * * * * * * * * * * * VD ..

C 0 en t- ,. on

N 00

N .

I IC CD

'CC ,-. CD CD CD CD CD CD CD .. C4 C4 . . . . . . . . .

* * * * vn cV eh CD 00 Vl N .. CD WI N CD en co s.o ol M o 0 0 O O 0 c) el CD CD e. CD CD CD

s s

* * * * * * * * * * * * * *

M *N•1 00 cA c^ .. 00 N In on N N .. N CD CN CN C) .. C) .. et CD C) CD CD .. ,. •. c.:, ,. m c) c) . .

* * * * * * * * * * * * * V- e. C4 00 VD V- on en ct on ,-. el 4-4 CV 00 e. Ve en ,. -. ,. CD C) CD CD Cl CD ,. CD CD vsl ,. C) cD c) c . . . . ) 1 ) 1

* *

VD 00 Cl M CD VI et M CD N .. CN CD CD CD CD CD .. CD CD CD CD CD CD •.

" " 1 •

CD Cl VD et CD CD CD CD O

N

* * * * * * * * * * * * * * * * * * * * * * * * * * * eh CD 00 VD 00 et et CN VD CD en 00 CN N ON VI e. N VD en s-, st-s 0 ,-1 NN ,-1 0 ,-1rn ,--.1NO 0 N N 0 0 C> ■-1

. . . . . . . . . . . . . . . . . . .

* * * * * * * * * * 71C4 el- 411 O\ CN ON eh VD eV CD VD en V) In VD Cl

NI CD en (,) O Cp O c) C> c) O O O cx) c)

U-s es

M 0"' M y, 4-■ =

4" 0 er 0 ..- cu 0 t = 5 5

0 o --4-. I.' .3) .1.. ■••■ E 0 rif '''

P., FL) '1E 5 " u -4- lE 0." .39 e --a •-• •- = s ° ?.) = 6 •- -6 t4 -s; 5 '' ce w ° 0 .0 o.....--,,

a) o 6 o 1■3 E=.( ,, C].., to. 4.) 4.. con o E" &) = = = = 0 0 .

.... 4, M

0. f.., = 4-. aq > c.) o -' = --, -^, ... a) 6 • .41 0 .., ,., ▪ 0 , 0 - - c.) 0 0 = 6 .... -4- 6 6 ›

•

0 6 •- .0

4. 4. ,..,0 4g 5 o 6 E T-1 io lo O o I) Q. - ,n CU 0 w ...2 NM 0 M a.) a) Q. co a7 .4 0. az .0 74 s 4" -' > > > to ..-, •.= 4., 1. .„ , 4., al .= -. 3-. i. ..= ,.., , ..) co

0 .1 r ,CI M M •M M -I-." al".. 54 I.." ..t54 g t ig Ct ai jro 72.

1:13 0.1 cp • • ■-• • 11.1 E = 0 '''' U 4 1 > < <1 <1 (... 14 1...1 CA 0 : V I■4 1■I PO PO PO CO)

_. c5 r4 e4 vi 4 vi .6 t4 00cl; c5 r4 e4 N M 4 Ili s.6 C.: 00 OWN 1.1 r. r. rd 1..1 1.0 1...1 rd 1.0r1 N N N

358

* * * * * * I yl yl

* * * * * * * * —

CP, m t')

* * * * * , * * * * I :) N

■0.

. .

* * * * * * *

••ch (-4

• •

* * * * * * * * *

I * kr) 00 ,.10 N ,t 00

.gt N

* * * * * * * * * * * * * * * * I cno0C---t---t--C)cnN I Cl kr1 e‘l"c1- • . .

* * * * * * * * * * * * * * * * * ten l() 01 ,C) Cc') "Zr Cn en In • • • • •

* * . * * * * * * * * * * * * * * * * * 00 v•-■ N M "3•

N N met d' • • • • • • •

O cr 0 0

el o .,.■ = .*, = 4'. %

•

) IA 1: ...E :- ti

. . m o cr L. .0.) E 4

° E .= ).- -c— E d E — E w,

* ct M a .°

E ,..„ 2 m E o ^0 gr ° ' 7-1 5 o 1: w o ,,, c-, E-1 4., CLI 4. C..)

g i 5 .E9 i ,; 15 ;!, T, t t 1: I; g g 2: ;, E „, 6 ,0 .41, 0 a.) .._ an m 6 W

C7

•

t wt 8 .°<T,E,E 51 m L-clEE 1:120 w .... ,, 1: 0.4 JD ; ,. .Z .. 11 - ,, Q. W w 4

w.,..,.. = 0 " , cc, OD > = 6 > > -. .> >. o. 13 bi) e,c, •-, •-t q.) 6 .7 el 16) 0: L. ,.. E 2 . c,.; r, .r, ,

g z 4) = = az m m m m a r.: g , t. to 5 a a ,,' ;-. m

! k 0 0.0 el.) co 0 • -. -- 0 5 -,a■,.(m1-.-4EEm: oo

i-. -eC (4 1. < < < C..7 )-1 0-4 cn C) : CY ■-■ 00 P.( A.i cf)

.-i N Mei vi vi, t• 06 O1;:4' ;;1 f2•1 MH *:. 4 4 ,2, 4 ,: A (., :...i

359

E O

O

0 0.)

O

nd CC

CC

O

44-4

tzt

0t: 44 O

O

7141

c.>

O

Tabl

e V

III

met

CC

O 1-0 1.

Jo

b T

enur

e

CP,

00

N

■C,

N

0.0

CN 00 0 CD

* * * * * * * *

N 01 N

.. . CI .. CD CD . CD CD . .• . .

* * * * *

I * * * * * C) un on r--

-

C) CN .4") CD CD cfl

.. C4 .. Cl cl c?

* * * * * * *

CD ,tC

CD C4 .. <I" 4-) CD .7i ,"2 cp cD

* * *

C) 0t N V-> C) ,f CN CD CD CD CD CD CD CD CD CD Cl CD CD . . . . . . . . . 11111111

* * * * * * * * ,..0 O.. kr.) .. ,z1• VI tc> CD .. r. 00 N In CD .71•0 CD CD CD CD CD CD CD CD CD .1 C4 .. Cl

. . . . . . . . . .

* * * *

Vl N \.D CD c+1 d- 01 vn N c, en 00 VD cn cn 0 O C> CD CD CD CD CD CD CD .. CD CD CD

. . 1 1

*

* cncf100 N C> N CD 0 0 0 C> . . . .

* * * * *

0 * * * * * * * N .. Cl 00 C.- Cl CD CN •,1• CN WI CD N N 00 ., S o'l

.. .. .. CD CD CD CD .. CD CD CD CD ci •-. c) c) c)

. . .

I 4-) 00 Cl LC> VI CD N 'cf N CN cD CJ C CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD ..

. . . . . . . . . . . . . . .

* * * * * * * * * * * * * * * * * * .ct• CD 00 10 00 .c• 00 N N CN .. 00 on CN CN V's) .. N %.0 01 .. V-) CD .. C4 C4 CD CD CD .. CD CD CD CD C4 CI CD 0 c) .. ,. • • • • • • • • • • • , . .•

* ..X. * * * * * * * * * * * * * * * * N N d- in or> CN CN 00 1D..- N .. CD on VD VD vl 1D N •-■ ,-■

.. .1 CD CD CD CD CD C9 ., CD cn CD M.1 .1 .. N CD CD CD ..

, . e .•

:114 tb = Cc,

pa 0 el., o q:

r a ..., iw = Cr 0 4., 0 -F, W

gg 5 c > 4: 4' C) C 4: -. W i-, M 0 CA 4' .. , 0

C '*' bk W +' r W

n cr 4. W C C m ;1-• 5. cl.)

O E "' - cp = 5 o W •W lw ;.. •E bmi • 474 E IA 1; gr, ...

M U i. O m

m„ E t4 E , 0 0 d... 1...

W 0 C4 • - E o . qz G 4.4 c, cr ..1 -4. 0. 4.

0

G 0

.0 G • . o E w 0.) 4) = = ,.. Etz. 0 .° > " 15 - . 0 d . cz 6 0 . = 14 q

• ,

z -i- c.' C 4.i 4wa = C. L.,

4' t g' . En 4-• 1') '"

4. 1. 2 0 gt, 5 m I: ,o o 0 g - . 0

0E

c4 „, ..) N 0 4.) CL1 0 G 42 4:=

"I: 74 P• • ■•■,, j. • y. ',. )" . 12.

; Er g 5 5 un

,c, IA F, . .„ 0

•

1. .... , 0 .= .. G 10 • W

6> G G 0 C = = 0. 6. = 'IN) „bp g gi. t t ,-1 =

WO W 0 0 ..-. ... 0 V ',1 (.7 •-4 )-. <1 <1 <1 (.. .5 ,.-5 col 0- : 0, .. •, pi p• co r4 -4-, •-• or. _. 0 ,-; ri ri 4 uri ■6 t--: 0(3 or; 6 .4 ei cn -x

N M .4 Lei v:; t.--: o0 01 I-. .. .1 .0 .0 r. H.0 ti r. N N N m

N CD CD CD

* *

* * * CD Ch CN

CD CD on CD CD

. . . 1

360

N N

N

O N

1-4

7-1

00

N

Ln

at

M rl

N

Tabl

e V

III

* *

m 71-

* * * * CN

I et cn . .

* * * * * *

7t VD Vi 7t

* *

1- •

* * VD

.

* * N 1.11

•

* *

vn •

* * cn

.

* * 41 41

•

CD N Cl 8 cP 0 cp (D (D.

* * * 00 CD 01 01 71" N CN

C). CD. C) CD. CD.

* * * * * * * * * * N 01 CD VD .41 CD CN VD

CI '1 (1 (1 ci c? cl

• * * * * on on 01 VD N 01 CD VD CD

C) 1-■ •-■ C:t esi • • • • - • • • •

* ** * * * * * * * * * * cn C In cn 01 71- cn cr) cn cn NM C) C) <I- on ,--, ,—, •-- ∎ .--i . .

4'

4' 73 = c

1-ct .^ or c c 0.) 0 := Cr ; +4 •1■■ 0

t' CP 5 0 > = no = , .. 4? I. --, e 0 .„ -, . ., -Et E -I m m = = E .' to

CP t■-■ 0 5 5 .3.) ::: u u cu o ° = E ., ., .t..t 6. e 5 0 e 0 m

4. 4. 0 E °

. ,.., at 'A CI CD ,..9 > i••• = 4.. '1 Gd

•

'= .5 ...5 tE) I-9 7:r.)c

t. 0 u c -t-t — , u = L. i. 5 = F.) cl.. L.

cu . 4. i.) 09 ' ;-, , 5 co $-, — -0 , , +7, ;.., L0 ,. . ,2° 44 '' Mt 5 5 cicu 7) a

• o o = . > > Fu ..7..■ CL• SI ir ot) -„F, ?,: t t, :te, v.r, .z, 6, o =

^0 bp w, -^. .--, 0 s... .. = a) az a) = = ,rz = = c a: 01. 01) pl) ., 6, I. u.,, 0. 0 dPel.) a) 0 .-, • ,-, -, a) 5 5,,, = 1„... 0 5 5 0: ,,,,

@ (..7 1-4 1— ..<1 <1 <I (.7 0. t—t cn 0 : C.' t—t I■1 PO 0.1 PO CA

(6 ,4 (-4 ri 4 vi v6 r-: 06 CT 0 4 (.4 ,4 (.4 ri 4 vi v6 r.--: at; cr;

361

00

E

I

1.

Job

Ten

ure

* 1/4"") N

* * * 00 •zi- 00 R

* * * * * * * * NM . . . .

I •--3 VD "71. R R O R . .

I I

I ONVDMN'zh 0 0 0 0 CD . . . . . .

*

I * 71- O

0

I

3,-3 3-3 0 I.

* VD (71 N lel 01 0 Le) k..0 C:) 0 en0 • 0 . . . . 0 0 .

* * * *

•,:t* M71- N 00 . . . .

* * * * In .71' VD N N •71'

R 0 0 0 .

0 .e.; it ..c CD 4-3 CD

E r

O. 7., 0., .3 ++ — s ue,

au .101

© a) E-1 C.: C.: :•-■ Cl.: 3-. C Ct '=.' = : = CI)

CA 03 CD 0: 0 (3.1 en I. ...4 MI 0I) • ■-■ I•■••■ -4-a p a) a) (:' c0 t t 0 ' 0: X. N C.: . :., O 30 .0 3.4 t ''' 4.3.

= 0 0: .0 = ,..1 11() Il 0 6,1 E-: CI) ..r:

CD = = az 40 C3.3 0.13 ti) = )c.z., ® C)

0.p a) 0) c:) o =

•

? .... CI) V. \i

', C.. 1-1 )—D 1— v) 0 = 44 P. 4-: r::* ali •3■3

C■i tri 4 :6 ■0: C• 00 C:3 ?-1 7:I c4f)

ca 4:

362

"1- CA VD CV C- I CD CD CD CD CD CD CD CD CD

* * * * * * * *

on N CD vl d' CI ,-. DO*

CD vn NN ol Op CV CD ,-. 0 ,-• . . . . . .

Job Te

nure

* * *

* * * * * vD .--4 00 N On N Cr) (Nn VD (,1 N ,-. CD CA cl ,1 ,1 CD ,-. C)

* * * * * *

* * * * * * *

N on c,n Lel 00 ..7c

N 0 Nt oo

,. ,. cl on cl on on ,. CD CD ,. CA . . . . . . . . . . .

N N Vn 'zt V1 flc) N VD 00

c.7 0 t Un

C) C) CD CD cD • • • • • • • • • • • • •

ctd• CD r- 00 Cl N Tr

CD CD CD CD CD CD ,-. CD CD CD CDCD

. . . .

* * * * * *

,-. d•,--. CD on VD N et c,1 7h N d" .. d'on 0 c) cp -. 0 0 0 ,. 0 cq CD CV CD CD

1 t• 1 1 1

* *

M 00 CD on

r- c? '1

VD VD N on N C- c) O CD CD cp O

M ". CD CD

** * * * *

rn c,n N 00 ,• "I- VD 71' .. cn •:• 7• t'n N . . . . . 1 1

. . . . 1 1 . . 1

* * * * * 0 ,-. Vn N If \..D ,-. '.7t N ,t CD tin N ,. N 00 N Vn cn CD cfn CD CD CD • --1 CD CD CD ,-1 •CD CD ,.. 0 ,. c) O. 1 1 1 1 1 1 1 1

O •-

a) .). c

a) 0 --, .. t). -4-4 0

= = = a) -4- a) ..,.

c 10.0 i = . 0 =

t-. 0

E =

.. W . CWC . E. g I 8 ,-, el eel •1. If) 01.1 iT, y., CD CD CO 0 0 .. E 0) 44 4-, 0000 0

0

`..4 0 10 C 0 t.-1 0 0 0 0 E-1 cp rA t W &) .. = ›. CC 74 = . ;. W 0 /. 1. I. 1. 1. 44 0 0 a; 0

Cr)Cl)

c lit >a, Co 01.1 C. 01. Po .. m 6XXXMO.S a -, a a: „,,, _ CD CD CD 0 CU

0 CD ,,, 0 1. w cn Cl) Cl) 1? - W at, 71. • "' 7 7 7 7 7

a) •,-, t .7." 16 0 .„,-, it. 0 ,t, 0 0 0 .2 .2 . 5-' . 5-'

• • 0 = 0 .0 .0 CL WI) tf) al. a) a) a) a) a) al) a) a) 0 0 c 1. .-, , C3 a) C).) c 1. 1. 1. I. 7.

'kt C, )'t in CA 0 Z F.1 00 PO '.1 CA ad PO al. a. a.

CS ,4 e4 eri Ti IA vi r-: °C. c.∎ ,4 e4 re; 4 46 vi. r4 06 cA

= Ut

Cd

CC

O E

un

• ■■c

un co) 4>

nr

un cr)

c•*1

CD E CC

O

Lr i. O Ca) it O.)

I■1

363

N

* * * *

I C-I en

* * * * * * * * CI \CD 0 0 tcl en N

I M Cr) kr) c, O • • • • •

* * * *

I N ch 471* Id Cl

* * * * * * C 00 rn M kn

'f?

* * * , * * * *

rn •—■ r-- N 0 0

. . . . . .

O • _1

= c>.) ).

C la'°) a) o ••-• •-•• 4-■ ,-` ,

= c#2 = = 4.1 4-' 6) ..... a) o 0.0 ? =

= 4-' a) a) 4-Z .17.: o = = E .4.... . E 8 es, n .Tp In

...., , 7:1 ev .-4 %,_, , 0 sv 4) 4., ...., , 0 Q c..) e.) r..) 6) 1.7)

6) d) t.1 = = = 0 C 4... .

,e 6) . GO L. vi 6) 6) 0..) CL.1 04

& 2 .-9 41 .C2 1:0) ..Ctl.) .E:o.> i . t.) at c 3... O. = -I-. = O. O. 571.1 04 CL 7... ).' C) 'D Mheih * = 442 ›' i< PI >< x x a) rk we'ller)"'7-1'71- 7-1 . ca L'a)a)a)a)a) o a: ).. a., , ,•-■ •-• o

W d N c4 O 6 .4 m c/1 m h.1.

E 11) 46:0P g &) (E" On4o7.=—C4c0000 O o o = = m o Cs. o L.,

1-D C.) )4 )•" cl) 0 *-■ P.4 (A a. a a a. a c■ O . 7-4 V■ ■:):1

364

00

I " N N

1-4

* * 00 0, N 0

1—■

d .

* * * *

CO )e) on •Tt.

* * •r)

N 0

C

N 0

* * 0 S en 0

. .

M en 0 0

oo

1-1

—

0

O■ 0 .

0 0

00

a) 2 >, C o L _,-. 0

4 O 0 ..= o. r.,•

0

> -H c

•

.c _ e.) 1... 6. . 4... ›- .1) ft 5 on 4.) a.'

o L., 1:1 C 0 4 4"

o' 4... .1■

o . -4 -I

a)

CI1 I. , .

C C CI cn tn = .4-■ CL) C 0 c.) W W a) a) O 0 >

o '

o a) > cu 5 o.) 5 L 0.) .0

O o C c cu 4:: E 0. -4-' :c. CC....• {..) -4-' :- c., 4•-• •- , 0.• •g 0 -., 0 -1-, •••• = ',1 = 0 O. 01., ).. = .= 5 = -,5 o ,•-• o 4, a) -,-, u c..1 4) L al 0)

C el -4-' L. ") L.. c.. c.) L.. 44)

.

44c>44C a) ow > C. g ›'„g:2' '71 ›"' = .c1 4,, 4.)

O 1-' 0 = a-). . '-' L o

cl) ca 0 a) ,,,, , = a' .C3 >4 X CI =

..1 • 0+ C.4 4 "4 4 4' 0 ° 4' • ,. O "r, 0 ›.r • ., 7.. VI r-I

4. a; c,.) 0 ,L) . • ,., 0 0 0 = z c). cl...z ra, 0 . 1).4 C:' ,•-■ et ). .0-3 0 0 0 0 '.0 .4) = .0 0.) 40 V. W (0. F4 7. VI cn gal z .r.' 0.) Ft E-q 4.) = 0.1 0. •.. cn 4:

\D I —

* * I en <1. (.1

* •ze

0 OM . . .

* (,) co u-) 71:

r:1 eei er tri o r aci COD *

365

00

* 4

5 * •- 1/40

* 1 CO OC, v--■ -4-1/4 N O. -1 O ' cf) C.) z *

* CI ' V7

•71- 00 ■D (7, 1 ci 71: N —: „y

s-■ CI

* * * *

1.) 1 lID 0 \ In ‹:/ 00 7:3 'el. ,--, '1/41- N 0 0 . . . 0 5

4-4 * * * * * *

0 0 N 1 M ,z)l• 7f- 0 t--- cr,

oo kr) M• , M . . . . .

0 Cr O C Cr) cl- O∎ N M M--, .4.) Ni 1 0 OM 0 0 0 0 . . .

P CD AI i .b. • ,(

* it tr) 1/40)mma,Nrn ■0 ..4 1 .7t

,_, *-. N ON --, --. . . . . . .

ct E E-I 0 +_ ,.,.> =

;-, . ... O -5

•- ..,_ . t .4, t _ 4,1 Ost) 71 -.a' al., = •- ,.... CI) 0 -4-. tt 4.4 = O C 1.) _ E el.) o c0 c E ..-..1 e. .

..0 =I 5 ° c. c 0 o ›- c o..)

*.r., ,,, •,, ›, . = ,F) 7...

▪

i... , >. .... Ctt 0 0 _,•.-,

... WI . LI) lo cp

749 <Id > > = 4.. ;.I I. - 0. cu ...

=

4. 0

OC. CL 0 P., ..0

0 1—' 4' a+I •4■, C.) V0I (01 -1•••

;••I 41. 4. = el' .- 0 .0

'5 = = -a .17'4 @I 4) -4.■ C.? O 0 ..-1 C. r6. '.. 1..1 0 0 7; 5 0 cs. a. = .— E 4-1 .4., c E,--, eu a) ..4

;., 6, c.i el CJ I. C CU CU r=1„, .. y_ a) at, fat) C. , 0 .: Xi

) 0 0 ;- C a) a a) a) .) E.) z a ,,,,,, ,y x >, p., ..... ....1 C • na

; a>. E5 I.0 I-. E, a a.) CI) cel ) ,0 cc

E E g'' g' ',9 0., 6 CU CU

PZ1 PA V) C4 0 Ci) AM = a) g E.. ..0

1/44 (-4 efi 16 06 a;

inte

rven

tion

366

•ci• It' CA M 00 * * *

8 E5 0 N 0 C> . . . . CO

"Ci 0

El

CD

* * * * * * * * * * * * * *

, \D ,--, ON ,--, s.17 CO N IC) 0 In t- CI.) 1 "1- c) cD -. -. •-, N ,--■ N ,--1

I

* *

I )r) M .

* * N M .

•cl- 0 .

)--. d' CD 0 1 I

0 0 0 ,--,

I'

Tab

le X

III

a ,J.,, * * * * *

* * * * * *

, cr, c.., 0. ,,, _ .., a .s , a, 0., in O in 1 ., _ c:, . . — "zi.. — , ., CD N • • • • • • • 6 ,

5 * * * * * 1 N )--, N 00SM ,..0‘.000 ,--, C•••••:1- )--,

01 1 w. r■I ,. C CD C) C.) CD CD CD CD 0 0

. . 0 CI 5 ;..., i * CD en V-) 00 N In m C) ,,:t CD In ,.0 .:I- .. N N

CD C7 CD CD CD ,-, CD ,-. CD CD CD C7 ,-. CD C••I 1 0 0.I a: * * * * * * * * el..) N * * * * * * * * * *

4 el 1 ,e CD 00 ■ID 00 ,:t Le) ‹) "Ce ON .. N ■.° on N .-... in 0 --, c■I c,1 -. . ,- --. 0 0 0 0 ,-, 0 .

* * * * * * * * * * * * * * * Cd)

▪

.6) * O CI N CI .:I- I4 cn (7) a, .--, .,1- .--i CT kk.0N •--, ■-■

* cn

4 . . . ■ . .' .* . . 1-1 I Ce1 0. C el ,-. ,--, ,-. el C:) C, 0 e::) 0 c,1 •--, 0 •—•

at • .-.1

CI = c

IPA o o ... .4 I:

=g am g _,. o . 5

> P' 4.9

s... a..

C3 = II-I: Ecy) ril .0 0 0 .,.. ›-

0 —

— (1) A) 4. 0. •"I a: .0 > n' 0 E 0

VS 0" 4. 0 -1. 0 5 0 0.) ° . Q c

c. c .

P. [.. I,

4-■ ,,_. ..w P. 4. O c y o . ...

U .,... .) +, • as al "' m ;.. 4j E: ,...

cs., >- .44 a) 0 az m , o

▪

w -4,-) l7 CO)

▪

II' = O 1-, 5 C >,

•

,.;.., a 0 E '4--; -0 -0 0 pt, .) cL)

• ô II' F .1 .E w E , Q „I .... a'= CO . CO = CO c.,.. a. a. 12. s... itr) 0 , )71) C.) .17 , .... •,-, • P.1 ..+ 0 = 4.3 1--I = , ;

b, '''t C) .4 <1 <1 <1 4 (... FSI Po Po Pi Un Pi • V

.2 6 i4 ci efi 4 urt, c: ,—, a

'..iii V RI

,4 N M4 tri l0 t-: 06 V A "0 1.1 1.4 i■I r. r. /1 1.. VD it el * *

367

* * * * * * , tin

VI rn

N • •

* * * * * * * *

en — ■.r) ",4" • • •

* * * * * * * * 100 cr■ a> 'a- en

N c's1 •-■ en • • •

* * * * * * * * * * * * d el en en CV rn • •

* * * * * * * * * * * * * * 0 00 rn 1/40 0 '.0 'r?N N ci (1

* * * * * * * * * * * * * * * 10 en CV -cr vl N et VI en "1" CV en ',C1- . . . . . . . .

O 0 O .° O

•

0 0 ° .."' >. > 0 C.)

•

0 E 0 -4.w '''

.0 0 0 9 >.)

.... "= 0 0 0 1..

■ <3.) 0 .4., O.

CI 0 > 12 0 E = -4.• t a) a) ....,

O +a C E 0 8

0

a>

(

•

-0 •".

•

;_. es CO , ■•• • 0 L. 0 =

O c: ^ E."( '^ e...) .... 0 0 0 ''' Lrz 0 ..0.. • ;• • • .a' . > *, ''' 3 .... el> ;••■ c to-, L. 0 4.)

CD 0 ■ 1 ' 4..) c, 4) = a) C. a) et .47.; = -4-. , = rz., s.) s..,

co)

=

•

l e.); . s .t4"

• 4-1

• "0 VO c4 P 0> S. z.,- El E, cu 8 ,—i 44 = = .0 0 ,C C 0 0... 0 O. 61 6, 6, al 6. O tk a, 40 0 • 0 E. ca E ea a) a) a a) I-I p, 7

In 0

1--) <4 (.7 •-; ■-.) <1 <4 <1 E-1 ■-■ C.7 )-• 0.4 Alo A-t GOO a 1 * ,r V 41; V 1:1.1 ,:, (.4 ei. ,(1: 0,3 vi t.: 00 ,:, ,?., 7:: ‘.:ii ,..; 7; 4 4 ,_, 66 10d

es •X *

368

APPENDIX 4: Cross Lagged Regressions (Site 1)

TIME 2

TIME 1 Sat. with Coll.

Trust in

Coll. Quality aware

Sup. partic. style

Mgt. commit

to quality

Team orientati

on

Job Tenure Age Gender Length of service Job Title 2 -.18** Job Title 3 Satisfaction with colleagues .17** Trust in colleagues .24** ,53*** Quality awareness .58*** .14** Sup. participative style .16** . 38*** .30*** Mgt. commitment to quality .51*** Team orientation .44***

**= p<.05 ***--p<.01

369

APPENDIX 5: Cross Lagged Regressions (Site 2)

TIME 2

TIME 1

Sat. with Coll.

Trust in

Coll, Quality aware

Sup. partic. style

Mgt. commit

to quality

Team orientati

on

Job Tenure Age Gender Length of service Job Title 2 -.14 * -.14** Job Title 3 -.20** -.15** Satisfaction with colleagues Trust in colleagues .36*** Quality awareness -.20** . 57***

Sup. participative style .23** Mgt. commitment to quality .55*** .20*** Team orientation .26***

**= p<.05 ***=p<.01

370

APPENDIX 6: Factor analysis of items measuring team orientation and trust in colleagues (site 1)

Item

Factor 1 Factor 2


. 800 .18


. 790 -.02


.79° .27


.77° .34


.71° .31


.00 .84°

I feel I am really part of my workgroup .26 .76°

The people in my work group encourage each other to work as a team

.47 .56°


.46 .46°


°indicates factor on which item loads most highly

371

APPENDIX 7: Factor analysis of items measuring team orientation and trust in colleagues (site 2)

Item

Factor 1 Factor 2


.89° .05


.88. .04


.76° .33


.74° .32


.63° .21


.41° .38


-.09 .88°

The people in my work group encourage each other to work as a team

.54 .61°

I feel I am really part of my workgroup .42 .55°


°indicates factor on which item loads most highly

372

APPENDIX 8: Factor analysis of items measuring management commitment to quality and improvement in commitment to quality at the site (site 1)

Item

Factor 1 Factor 2

Compared to a year ago, in general, people at the site:

Exert greater effort in looking for ways to prevent mistakes .91° .19

Put greater effort into thinking about how they can improve their work .89° .17

Take greater pride in knowing they had made a contribution toward improving things around here .86° .21

Are more quality conscious .84° .23

Are more willing to put forward ideas and suggestions without expecting extra reward .73° .15

Management provides support for quality improvements throughout the organization .20 .86°

Management has attempted to involve everyone in continuous improvement .09 .79°

Management is genuinely committed to improving quality .27 .75 °

Management does its best to provide employees with the right tools and materials to do a quality job .08 .74°

Management sets examples of quality performance in their daily activities .33 .72°


° indicates factor on which item loads most highly

373

APPENDIX 9: Factor analysis of items measuring management commitment to quality and improvement in commitment to quality at the site (site 2)

Item

Factor 1 Factor 2

Management provides support for quality improvements throughout the organization .86° .12

Management is genuinely committed to improving quality .78° .15

Management has attempted to involve everyone in continuous improvement .73° .20

Management sets examples of quality performance in their daily activities .70° .22

Management does its best to provide employees with the right tools and materials to do a quality job .70° .08

Compared to a year ago, in general, people at the site:

Put greater effort into thinking about how they can improve their work .13 .87°

Exert greater effort in looking for ways to prevent mistakes .11 .86°

Take greater pride in knowing they had made a contribution toward improving things around here .20 .83°

Are more willing to put forward ideas and suggestions without expecting extra reward .09 .63°

Are more quality conscious .38 .57°



374

APPENDIX 10: Factor analysis of items measuring commitment to improvement (SITE 1)

Item

Factor 1 Factor 2

Strongly committed to Total Quality .81° .08

Contribution to improving things would please me .70° .15

Put a lot of effort into thinking about how I can improve my work .70° .28

Always looking for ways to prevent mistakes .65° .25

Looking for ways to improve how things are done is part of my job .16 . 790

I am not paid to think of ways of improving thingst .11 . 770




375

APPENDIX 11: Factor analysis of items measuring commitment to improvement (SITE 2)

Item

Factor 1 Factor 2

Strongly committed to Total Quality .77. .00

Contribution to improving things would please me .73° .26

Always looking for ways to prevent mistakes .71° .16

Put a lot of effort into thinking about how I can improve my work .55° .47


I am not paid to think of ways of improving thingst .17 .70°




376

APPENDIX 12: Factor analysis of items measuring intrinsic motivation and commitment to improvement (SITE 1)

Item

Factor 1 Factor 2

Feel unhappy when work is not up to usual standard .77° .09

Take pride in doing my job as well as I can .75° .17

Feel a sense of personal satisfaction when I do the job well .75° .31

Like to look back on the day's work with a sense of a job well done .74° .26

Try to think of ways of doing my job effectively .70° .31

Opinion of myself goes down when I do this job badly .64° -.18

Put a lot of effort into thinking about how I can improve my work .26 .67°

Looking for ways to improve how thngs are done is part of my job .10 .65°

I am not paid to think of ways of improving things j' -.05 .65°

I often put foward ideas and suggestion without expecting extra reward .05 .64°

Always looking for ways to prevent mistakes .36 .53°

Strongly committed to Total Quality .43 .52°

Contribution to improving things would please me .42 .49°


indicates factor on which item loads most highly t item reversed scored

377

APPENDIX 13: Factor analysis of items measuring intrinsic motivation and commitment to improvement (SITE 2)

Item

Factor 1 Factor 2 Factor 3

Feel unhappy when work is not up to usual standard .83° .09 .08

Take pride in doing my job as well as I can .77° .03 .34

Try to think of ways of doing my job effectively .76° .26 .18

Feel a sense of personal satisfaction when I do the job well .76° .10 .25

Opinion of myself goes down when I do this job badly .72° .09 -.01

Like to look back on the day's work with a sense of a job well done .56° .09 .55

Always looking for ways to prevent mistakes .43° .36 .19

Looking for ways to improve how things are done is part of my job .11 .79. .23

I often put foward ideas and suggestion without expecting extra reward .13 .73° -.15

I am not paid to think of ways of improving thingst -.06 .63° .40

Put a lot of effort into thinking about how I can improve my work .45 57° .12

Strongly committed to Total Quality .15 .08 .84°

Contribution to improving things would please me .37 .34 .53 °

Eigenvalue 5.29 1.61 1.00 Percent of variance 40.7 12.4 7.7

indicates factor on which item loads most highly f item reversed scored

378

APPENDIX 14: Predictors of intrinsic motivation and higher order need strength

Intrinsic Motivation Higher Order Need Strength

Predictor variablest Site 1 Site 2 Site 1 Site 2

Management commitment to quality Supervisory commitment to quality .15** Quality awareness .42*** .35*** .38*** . 39***

Organizational commitment .26*** .31*** .20*** .16** Intrinsic job satisfaction An improvement in quality climate An improvement in commitment to quality -.24***

**= p<.05

***--p<.01

t control variables are included but not shown here

379

APPENDIX 15: Predictors of employee participation using logistic regression

Predictors (Time 1)

Time 2

Participation in TQM Participation in TQM

Job Tenure .01 .03 Age .03 .03 Gender -.42 -.38 Length of service -.05** -.05** Job Title 2 1.06+ .90 Job Title 3 1.56** 1.42* Supervisor participative style .71*** .68*** Organizational commitment -.04 -.21 Higher order need strength .22 .13 Influence gap .06 .06

Perceived benefit of TQM interventiont

N 164

,,,,/,-;'- ,1//, 4:;51:,A, 73***

164

**= p<.05

t Measured at time 2

380

APPENDIX 16: Factor analysis of all items measuring improvement

Item Factor 1 Factor 2 Factor 3

There is greater contact between management and employees .77. .27 .22

People are encourage more to say how they think things could be done better .75° .25 .22

Visible progress has been made in improving things at the site .75° .19 .16

The level of cooperation between management and employees has improved .74° .30 .36

Top management is more supportive of suggestions to improve the way things are done around here .74° .23 .30

Top management is more committed to Total Quality .72° .33 .06

Communications between management and employees has improved .71 ° .28 .31

Total Quality is a greater priority at this site .69° .37 .02

People at the site:

Put greater effort into thinking about how they can improve their work .23 .87. .14

Exert greater effort into looking for ways to prevent mistakes .25 .86° .15

Take greater pride in knowing they had made a contribution toward improving things around here .36 .81 ° .14

Are more quality conscious .38 .73 ° .07

Are more willing to put forward ideas and suggestions without expecting extra reward .22 .64° .19

My performance on the job has improved .05 .12 .85°

The performance of my work area has improved .34 .21 .72°

Improvements have been made in how I do my job .40 .11 .71°

Eigenvalue 8.37 1.62 1.20 Percent of variance 52.3 10.2 7.5


381

BIBLIOGRAPHY

Ahlbrandt, R., Leana, C. and Murrell, A. (1992) Employee involvement programmes

improve corporate performance, Long Range Planning, Vol 25, No 5, pp. 91-98.

Armenakis, A.A., Harris, S.G. and Mossholder, K.W. (1993) Creating readiness for

organizational change, Human Relations, Vol 46, No 6, pp. 681-703.

Baum, B.H., Sorensen, P.F., Place, W.S. (1970) The effect of managerial training on

organizational control: An experimental study, Organizational Behaviour and Human

Performance, Vol 5, pp. 170-182.

Bennis, W.G. (1977) Bureaucracy and Social Change: An anatomy of a training failure,

in P.H. Mirvis and D.N. Berg (Eds.), Failures in Organization Development and

Change. New York: Wiley & Sons. pp. 191-215.

Bennis, W., and Nanus, B. (1985) Leaders. New York: Harper and Row.

Berlinger, L.R., Glick, W.H., and Rodgers, R.C. (1988) Job enrichment and

performance improvement, in J.P. Campbell and R.J. Campbell (Eds.), Productivity in

Oganizations: New Perspectives from Industrial and Organizational Psychology. San

Franciso: Jossey Bass. pp. 219-254.

Blackburn, R., and Rosen, B. (1993) Total quality and human resources management:

lessons learned from Baldrige Award-winning companies, Academy of Management

Executive, Vol 7, No 3, pp. 49-66.

Blake, R.R., and Mouton, J.S. (1964) The Managerial Grid. Houston TX. : Gulf

Publishing.

382

Boje, D.M. and Winsor, R.D. (1993) The ressurrection of Taylorism: Total Quality

Management's hidden agenda, Journal of Organizational Change Management, Vol 6,

No 4, pp. 57-70.

Bowen, D.E. and Lawler, E.E. (1992) Total quality-oriented human resource

management, Organizational Dynamics, Vol 20, No 4, pp. 29-41.

Boyle, R. and Bovacich, P. (1970) The development of trust and mistrust in mixed-

motive games, Sociometry, Vol 33, pp. 123-139.

Brief, A.P., and Motowidlo, S.J. (1986) Prosocial organizational behaviours, Academy

of Management Review, Vol 11, No 4, pp. 710-725.

Brocka, B., and Brocka, M.S. (1992) Quality Management: Implementing the Best

Ideas of the Masters. Homewood, IL: Irwin.

Brooks, M.P. (1971) The community action program as a setting for applied research,

in F.G. Caro (Ed.), Readings in Evaluation Research. New York: Russell Sage

Foundation. pp. 53-62.

Brown, R.J., and Williams, J.A. (1984) Group identification: The same thing to all

people?, Human Relations, Vol 37, pp. 547-564.

Bruning, N. S., and Liverpool, P. R. (1993) Membership in Quality Circles and

participation in decision making, The Journal of Applied Behavioral Science, Vol 29,

No 1, pp. 76-95.

Bryman, A. (1989) Research Methods and Organization Studies. London: Routledge.

Burawoy, M. (1979) Manufacturing Consent. Chicago: University of Chicago Press.

383

Burke, W. (1986) Leadership as empowering others, in S. Srivastra (Ed.), Executive

Power. San Francisco: Jossey-Bass. pp. 51-77.

Burns, J. M. (1978) Leadership. New York: Harper & Row.

Bushe, G. R. (1988) Cultural contradictions of statistical process control in American

manufacturing organizations, Journal of Management, Vol 14, pp. 19-31.

Campbell, D.T., and Stanley, J.0 (1966) Experimental and Quasi-Experimental Design

for Research. Chicago: Rand-McNally.

Campbell, J.P. (1990) Modeling the performance prediction problem in industrial and

organizational psychology, in M.D. Dunnette and L.M. Hough (Eds.), Handbook of

Industrial and Organizational Psychology. (2nd edition, Vol 1) Palo Alto, CA:

Consulting Psychologists Press. pp. 687-732.

Campbell, J.P., Dunnette, M.D., Lawler, E.E. and Weick, K.E. (1970) Managerial

Behaviour, Performance, and Effectiveness. New York: McGraw-Hill.

Chin, R. and Benne, K. (1976) General strategies for affecting change in human

systems, in W. Bennis, K. Benne, R. Chin and K. Corey (Eds.), The Planning of

Change (3rd edition). New York: Holt, Rinehart & Winston. pp. 22-45.

Clement, R.W. (1978) An empirical test of the hierarchy theory of training evaluation.

Unpublished doctoral thesis, Michigan State University.

Coch, L. and French, J.R.P. (1948) Overcoming resistance to change, Human Relations,

Vol 1, pp. 512-532.

384

Cole, R.E. (1980) Work, Mobility, and Participation: A comparative study of American

and Japanese industry. Berkeley: University of California Press.

Collard, R. and Dale, B. (1989) Quality circles, in K. Sisson (Ed.), Personnel

Management in Britain. Oxford: Basil Blackwell.

Conger, J.A. and Kanungo, R.N. (1987) Toward a behavioral theory of charismatic

leadership in organizational settings, Academy of Management Review, Vol 12, pp.

637-647.

Conger, J.A., and Kanungo, R.N. (1988) The empowerment process: integrating theory

and practice, Academy of Management Review, Vol 13, No 3, pp. 471-482.

Cook, J.D., Hepworth, S.J., Wall, T.D., and Wan -, P.B. (1981) The Experience of Work:

A Compendium and Review of 249 Measures and their use. London: Academic Press.

Cook, J., and Wall, T. (1980) New work attitude measures of trust, organizational

commitment and personal need non-fulfilment, Journal of Occupational Psychology,

Vol 53, pp. 39-52.

Cotton, J.L., Vollrath, D.A., Froggatt, K.L. Lengnick-Hall, M.L., and Jennings, K.R.

(1988) Employee participation: diverse forms and different outcomes, Academy of

Management Review, Vol 13, No 1, pp. 8-22.

Crosby, P. B. (1979) Quality is Free. New York: McGraw-Hill.

Crosby, P.B. (1986) Quality without Tears. New York: McGraw-Hill.

385

Cruise O'Brien R. (1995) Employee involvement in performance improvement: a

consideration of tacit knowledge, commitment and trust, Employee Relations, Vol 17,

No 3, pp.110-120.

Cruise 0' Brien, R. and Voss, C. (1992) In search of quality, Operations management

paper 92/02, London Business School working paper.

Cummings, L.L. (1975) Assessing the Graen/Cashman model and comparing it to other

approaches, in J.G. Hunt and L.L. Larson (Eds.), Leadership Frontiers. Kent, Ohio:

Comparative Administration Research Institute, Kent State University. pp. 181-185.

Cummings, T.G., Molloy, E.S. and Glen, R. (1977) A methodological critique of fifty-

eight selected work experiments, Human Relations, Vol 30, No 8, pp. 675-708.

Cutcher-Gershenfeld, J.A. (1986) A Workforce Divided: Theoretical and Practical

Implications of In-Groups and Out-Groups in QWL Initiatives, Working paper, MIT

Sloan School of Management, Cambridge, Massachusetts.

Dale, B., and Cooper, C. (1992) Total Quality and Human Resources: An Executive

Guide. Oxford: Blackwell.

Dansereau, F., Graen, G. and Haga, W.J. (1975) A vertical dyad linkage approach to

leadership within formal organizations: A longitudinal investigation of the role-making

process, Organizational Behavior and Human Performance, Vol 13, pp. 46 -78.

Dawson, P. (1995) Managing quality in the multi -cultural workplace, in A. Wilkinson

and H. Willmott (Eds.), Making Quality Critical: New Perspectives on Organizational

Change. London: Routledge. pp. 173-193.

386

Dean, J.W., and Bowen, D.E. (1994) Management theory and total quality: Improving

research and practice through theory development, Academy of Management Review,

Vol 19, No 3, pp. 392-418.

Decker, P. (1982) The enhancement of behavior modeling training of supervisory skills

by the inclusion of retention processes, Personnel Psychology, Vol 35, pp.323-331.

DeCotiis, T.A. and Summers, T.P. (1987) A path analysis of a model of the antecedents

and consequences of organizational commitment, Human Relations, Vol 40, No 7, pp.

445-470.

Deery, S.J., Erwin, P.J. and Iverson, R.D. (1995) Trade union instrumentality, dual

allegiance and organizational effectiveness. Paper presented at the International

Industrial Relations Association HRM study group, Washington, June.

Deming, W. (1986) Out of the Crisis. Massachusetts Institute of Technology, Center for

Advanced Engineering Study. Cambridge, MA: Cambridge University Press.

Deutsch, M. (1962) Cooperation and trust: some theoretical notes, in M.R. Jones,

Nebraska Symposium on Motivation. Lincoln, Nebraska: University of Nebraska Press.

Drummond, H. and Chell, E. (1992) Should organizations pay for quality? Personnel

Review, Vol 21, No 4, pp. 3-11.

Eden, D. (1984) Self-fulfilling prophecy as a management tool : Harnessing pygmalion,

Academy of Management Review, Vol 9, No 1, pp. 64-73.

Eisenberger, R., Fasolo, P., and Davis-LaMastro, V. (1990) Perceived organizational

support and employee diligence, commitment, and innovation, Journal of Applied

Psychology. Vol 75, No 1, pp. 51-59.

387

Eisman, R. (1992) Why companies are turning to total quality, Incentive, Vol 166, No

5, pp. 22-26.

Emery, F.E., and Trist, E. (1969) Socio-technical systems, in F.E. Emery (Ed.), Systems

Thinking. Harmondsworth: Penguin.

Ernst & Young and the American Quality Foundation. (1992) Best Practices Report.

New York: Ernst & Young.

Feigenbaum, A.V. (1983) Total Quality Control. 3rd edition. New York: McGraw-Hill.

Fiedler, F.E., and Mahar, L. (1979) The effectiveness of contingency model training . A

review of the validation of Leader Match, Personnel Psychology, Vol 32, pp. 45-62.

Fiman, B.G. (1973) An investigation of the relationships among supervisory attitudes,

behaviours, and outputs: an examination of McGregor's Theory Y, Personnel

Psychology, Vol 26, pp. 95-105.

Fisher, C.D. (1978) The effects of personal control, competence, and extrinsic reward

systems on intrinsic motivation, Organizational Behavior and Human Performance,

Vol 21, pp. 273-288.

Fleishman, E. (1955) Leadership climate, Human Relations training and supervisory

behaviour, Personnel Psychology, Vol 6, pp. 205-222.

Gevritz, C. (1991) The fundamentals of advanced planning, Quality Progress, April,

pp. 49-51.

388

Gist, M.E. (1987) Self-efficacy: Implications for organizational behaviour and human

resource management, Academy of Management Review, Vol 12, pp. 472-485.

Golembiewski, R.T, Billingsley, K. and Yeager, S. (1976) Measuring change and

persistence in human affairs: Types of change generated by OD designs, Journal of

Applied Behavioral Science, Vol 12, No 2, pp.133-157.

Golembiewski, R. T., and McConkie, M. (1975) The centrality of interpersonal trust in

group processes, in C. L. Cooper (Ed.), Theories of Group Processes. New York: Wiley

and Sons.pp. 131-185.

Gordon, L.V. (1970) Measurement of bureaucratic orientation, Personnel Psychology,

Vol 23, pp. 1-23.

Graham, J.W., and Verma, A. (1991) Predictors and moderators of employee responses

to employee participation programs, Human Relations, Vol 44, No 6, pp. 551-568.

Grant, R.M., Shani, R., and Krishnan, R. (1994) TQM's Challenge to management

theory and practice, Sloan Management Review, Winter, pp. 25-35.

Griffin, R.W. (1983) Objective and social sources of information in task redesign: A

field experiment, Administrative Science Quarterly, Vol 28, pp. 184-200.

Griffin, R.W. (1988) Consequences of quality circles in an industrial setting: A

longitudinal assessment, Academy of Management Journal, Vol 31, No 2, pp. 338-358.

Grummitt, J. (1983) Team Briefing. London: Industrial Society.

389

Guest, D. (1984) Social psychology and organizational change, in M. Grunneberg and

T. Wall (Eds.), Social Psychology and Organizational Behaviour. London: Wiley &

Sons. pp. 183-221.

Guest, D. (1992a) Human Resource Management in the UK, in B. Towers (Ed.), A

Handbook of Human Resource Management. Oxford: Blackwell. pp. 3-26.

Guest, D.E. (1992b) Employee commitment and control, in J. Hartley and G.

Stephenson (Eds.), Employment Relations. Oxford: Blackwell. pp. 111-135.

Guest D. and Peccei, R. (1994) A test of the feasibility of changing organizational

commitment. Paper presented at the British Occupational Psychology Conference,

Brighton.

Guest, D., Peccei, R. and Thomas, A. (1993) The impact of employee involvement on

organizational commitment and 'them and us' attitudes, Industrial Relations Journal,

Vol 24, No 3, pp. 191-200.

Guzzo, R.A., Jette, R.D., and Katzell, R.A. (1985) The effects of psychologically based

intervention programs on worker productivity: A meta-analysis, Personnel Psychology,

Vol 38, pp. 275-291.

Hackman, J.R., and Oldham, G.R. (1980) Work Redesign. Reading, MA : Addison-

Wesley.

Hackman, J.R., Pearce, J.L., and Wolfe, J.C. (1978) Effects of changes in job

characteristics on work attitudes and behavior: A naturally occurring quasi-experiment,

Organizational Behavior and Human Performance, Vol 21, No 3, pp. 289-304.

390

Hand, H.H., and Slocum, J.W. (1970) Human Relations training for middle

management: A field experiment, Academy of Management Journal, Vol 13, pp. 403-

410.

Hand, H.H., and Slocum, J.W. (1972) A longitudinal study of the effects of a human

relations training program on managerial effectiveness, Journal of Applied Psychology,

Vol 56, No 5, pp. 412-417.

Hand, H.H., Richards, M.D., and Slocum, J.W. (1973) Organizational climate and the

effectiveness of a Human Relations training program, Academy of Management

Journal, Vol 16, pp. 185-195.

Head, T.C., Molleston, J.L., Sorensen, P.F., and Gargano, J. (1986) The impact of

implementing a Quality Circles intervention on employee task perceptions, Group and

Organization Studies, Vol 11, No 4, pp. 360-373.

Hespe, G., and Wall, T. (1976) The demand for participation among employees, Human

Relations, Vol 29, No 5, pp. 411-428.

Hill, S. (1991a) How do you manage a flexible firm? The total quality model, Work,

Employement & Society, Vol 5, No 3, pp. 397-415.

Hill, S. (1991b) Why Quality Circles failed but Total Quality Management might

succeed, British Journal of Industrial Relations, Vol 29, No 4, pp. 541-568.

Hill, S. (1995) From Quality Circles to Total Quality Management, in A. Wilkinson

and H. Willmott (Eds.), Making Quality Critical: New perspectives on organizational

change. London: Routledge. pp.33-53.

391

Hill, S., and Wilkinson, A. (1995) In search of TQM, Employee Relations, Vol 17, No

3, pp. 8-25.

House, R.J. (1968) Leadership training: Some dysfunctional consequences,

Administrative Science Quarterly, Vol 12, pp. 556-571.

Ishikawa, K. (1985) What is Total Quality Control? The Japanese Way. Englewood

Cliffs, New Jersey: Prentice-Hall.

Jacob, R. (1993) TQM: More than a dying trend?, Fortune, October 18, pp. 66-72.

Jacoby, J., and Terborg, J.R. (1975) 'Managerial Philosophies scale'. Texas:

Teleometrics International.

James, L.R., and Jones, A.P. (1974) Organizational climate: A review of theory and

research, Psychological Bulletin, Vol 81, pp. 1096-1112.

Jenkins, G.D. and Lawler, E.E. (1981) Impact of employee participation in pay plan

development, Organizational Behavior and Human Performance, Vol 28, No 1, pp.

111-128.

Jenkins, S., Noon, M., and Lucio, M.M. (1995) Negotiating quality: The case of TQM

in Royal Mail, Employee Relations, Vol 17, No 3, pp. 87-98.

Jermier, J.M., and Berkes, L.J. (1979) Leader behavior in a police command

bureaucracy: A closer look at the quasi-military model, Adminstrative Science

Quarterly, Vol 24, pp. 1-23.

392

Jones, A.P., and James, L.R. (1979) Psychological climate: Dimensions and

relationships of individual and aggregated work environment perceptions,

Organizational Behavior and Human Performance, Vol 23, pp. 201-250.

Juran, J. M. (1989) Juran on Leadership for Quality. New York: Free Press.

Kanter, R. M. (1968) Commitment and social organization: a study of commitment

mechanisms in utopian communities, American Sociological Review, Vol 33, No 4, pp.

499-517.

Kanter, R.M. (1988) When a thousand flowers bloom: Structural, collective, and social

conditions for innovation in organizations, in B.M. Staw and L.L. Cummings (Eds.)

Research in Organizational Behaviour (Vol 10). Greenwich, CT: JAI Press. pp. 169-

211.

Kanter, R.M. (1992) The future of bureaucracy and hierarchy in organizational theory:

A report from the field, in P.Bourdieu and P.Coleman (Eds.), Social Theory for a

Changing Society. Boulder, CO: Westview Press.

Katz, D. and Kahn, R.L. (1978) The Social Psychology of Organizations. 2nd edition.

New York : Wiley.

Kearney, A.T. (1992) Total Quality: time to take of the rose tinted spectacles, a report

in association with the TQM Magazine. Kempston: IFS publications.

Kelly, J., and Kelly, C. (1991) 'Them and us': social psychology and 'the new

Industrial Relations', British Journal of Industrial Relations, Vol 29, No 1, pp. 25-48.

Kelman, H.C. (1958) Compliance, identification and internalization: Three processes of

attitude change, Journal of Conflict Resolution, Vol 11, No 1, pp. 51-60.

393

Kerfoot, D. and Knights, D. (1995) Empowering the 'quality worker': The seduction

and contradiction of the total quality phenomenon, in A. Wilkinson and H. Willmott

(Eds.), Making Quality Critical: New Perspectives on Organizational Change.

London: Routledge. pp. 219-238.

Kiess-Moser, E. (1990) International perspectives on quality, Canadian Business

Review, Autumn, pp. 31-33.

Kimberly, J.R. (1979) Issues in the design of longitudinal organizational research, in

R.T. Mowday and R.M. Steers (Eds.), Research in Organizations: Issues and

Controversies. Santa Monica, CA: Goodyear. pp. 121-136.

Kirkpatrick, D.L. (1976) Evaluation of training, in R.L. Craig (Ed.), Training and

Development Handbook. 2nd Edition. New York: McGraw-Hill.

Klein, J. A. (1984) Why supervisors resist employee involvement, Harvard Business

Review, Sept-Oct, pp. 87-95.

Kohn, A. (1993) Why incentive plans cannot work, Harvard Business Review,

September- October, pp. 54-63.

Kozlowski, S.W.J., and Doherty, M.L. (1989) Integration of climate and leadership:

Examination of a neglected issue, Journal of Applied Psychology, Vol 74, No 4, pp.

546-553.

Latham, G.P. (1989) Behavioural approaches to the training and learning process, in

I.L. Goldstein (Ed.), Training and Development in Organizations. San Francisco:

Jossey-Bass.

394

Latham, G.P., Wexley, K.N. and Pursell, E.D. (1975) Training managers to minimize

rating errors in the observation of behavior, Journal of Applied Psychology, Vol 60, No

5, pp.550-555.

Lawler, E.E (1977) Adaptive experiments: An approach to Organizational Behavior

research, Academy of Management Review, Vol 2, pp. 576-585.

Lawler, E.E. (1982) Increasing worker involvement to enhance organizational

effectiveness, in P.S. Goodman (Ed.), Change in Organizations: New Perspectives on

Theory, Research and Practice. San Francisco: Jossey-Bass.

Lawler, E.E. (1994) Total Quality Management and employee involvement: Are they

compatible?, Academy of Management Executive, Vol 8, No 1, pp. 68-76.

Lawler, E.E. and Hackman, J.R. (1969) Impact of employee participation in the

development of pay incentive plans . A field experiment, Journal of Applied

Psychology, Vol 53, No 6, pp. 467-471.

Lawler, E.E., and Mohrmam, S.A. (1985) Quality circles after the fad, Harvard

Business Review, Vol 63, No 1, January-February, pp. 65-71.

Lewin, K. (1943) Defining the "field at a given time", Psychological Review, Vol 50,

pp. 292-310.

Lewin, K. (1951) Field Theory in the Social Sciences. New York: Harper.

Likert, R. (1961) New Patterns of Management. New York: McGraw-Hill.

Likert, R. (1967) The Human Organization : Its Management and Value. New York:

McGraw-Hill.

395

Lillrank, P., and Kano, N. (1989) Continuous Improvement: Quality Control Circles.

Ann Arbor, Mich: Center for Japanese Studies, University of Michigan.

Livingston, J. (1969) Pygmalion in management, Harvard Business Review, Vol 47, No

4, pp. 81-89.

Locke, E.A, Feren, D.B., McCaleb, V.M. Shaw, K.N., and Denny, A.T. (1980) The

relative effectiveness of four methods of motivating employee performance, in K.D.

Duncan, M.M. Grunneberg, and D. Wallis (Eds.), Changes in working life. London:

Wiley.

McArdle, L., Rowlinson, M., Proctor, S., Hassard, J., and Forrester, P. (1995)

Employee empowerment or the enhancement of exploitation, in A. Wilkinson and H.

Willmott (Eds.), Making Quality Critical: New Perspectives on Organizational

Change. London: Routledge. pp. 156-172.

McGregor, D. (1960) The Human Side of Enterprise. New York: McGraw-Hill.

McGregor, D. (1967) The Professional Manager. New York: McGraw-Hill.

Marchington, M. (1980) Responses to Participation at Work. Farnborough: Gower.

Marchington, M., Wilkinson, A., Ackers, P., and Goodman, J. (1994) Understanding

the meaning of participation: Views from the workplace, Human Relations, Vol 47, No

8, pp. 867-894.

Marks, L.M., Mirvis, P.H., Hackett, E.J., and Grady, J.F. (1986) Employee

participation in a Quality Circle program• impact on quality of work life, productivity,

and absenteeism, Journal of Applied Psychology, Vol 71, No 1, pp. 61-69.

396

Mathieu, J.E., and Zajac, D.M. (1990) A review and meta-analysis of the antecedents,

correlates and consequences of organizational commitment, Psychological Bulletin, Vol

108, No 2, pp. 171-194.

Meyerson, D. and Martin, J. (1987) Cultural change: An integration of three different

views, Journal of Management Studies, Vol 24, No 6, pp. 623-647.

Miller, K.I., and Monge, P.R. (1986) Participation, satisfaction, and productivity: A

meta-analytic review, Academy of Management Journal, Vol 29, No 4, pp. 727-753.

Mohrman., S.A., Tenkasi, R.V., Lawler, E.E. and Ledford, G.E. (1995) Total quality

management: Practice and outcomes in the largest US firms, Employee Relations, Vol

17, No 3, pp. 26-41.

Mowday, R.T., Steers, R.M., and Porter, L.W. (1979) The measurement of

organizational commitment, Journal of Vocational Behaviour, Vol 14, pp. 224-247.

Noe, R.A. (1986) 'Trainees' attributes and attitudes: Neglected influences on training

effectiveness, Academy of Management Review, Vol 11, No 4, pp. 736-749.

Noe, R.A., and Schmitt, N. (1986) The influence of trainee attitudes on training

effectiveness: Test of a model, Personnel Psychology, Vol 39, pp. 497-523.

Oakland, J. (1989) Total Quality Management. London: Heinemann.

Ogbonna, E. (1992-1993) Managing organizational culture: Fantasy or reality? Human

Resource Management Journal, Vol 3, No 2, pp. 42-54.

397

Olian, J.D. and Rynes, S.L. (1992) Making Total Quality work: aligning organizational

processes, performance measures, and stakeholders, Human Resource Management,

Vol 30, No 3, pp. 303-333.

Organ, D. W. (1988) Organizational Citizenship Behaviour: The Good Soldier

Syndrome. Lexington, MA: Lexington Books.

Parker, M., and Slaughter, J. (1993) Should the labour movement buy TQM?, Journal

of Organizational Change Management, Vol 6, No 4, pp. 43-56.

Peccei, R. and Guest, D. (1993) The dimensionality and stability of organizational

commitment. Centre for Economic Perfolinance Discussion Paper No. 149. The

London School of Economics.

Peccei, R. and Wood, S. (1994) The impact of Total Quality Management on quality

consciousness, mimeo, The London School of Economics.

Persico, J. (1991) Leadership and empowerment in a total quality organization, Total

Quality Management, Vol 2, No 1, pp. 57-67.

Pettigrew, A.M. (1990) Longitudinal field research on change: Theory and practice,

Organization Science, Vol 1, No 3, pp. 267-291.

Porras, J.I. and Berg, P.O. (1978) The impact of Organization Development, Academy

of Management Review, Vol 3, pp. 249-266.

Rafaeli, A. (1985) Quality circles and employee attitudes, Personnel Psychology, Vol

38, pp. 603-615.

398

Ramsay, H. (1976) Participation: The shop floor view, British Journal of Industrial

Relations, Vol 14, No 2, pp. 128-141.

Randall, D.M. (1990) The consequences of organizational commitment:

methodological investigation, Journal of Organizational Behavior, Vol 11, pp. 361-

378.

Rees, C. (1995) Quality management and HRM in the service industry: some case study

evidence, Employee Relations, Vol 17, No 3, pp. 99-109.

Reeves, C.A. and Bednar, D.A. (1994) Defining quality: alternatives and implications,

Academy of Management Review, Vol 19, No 3, pp. 419 -445.

Reger, R.K., Gustafson, L.T., Demarie, S.M., and Mullane, J.V. (1994) Reframing the

organization: Why implementing Total Quality is easier said than done, Academy of


Rhodes, S.R., and Steers, R.M. (1981) Conventional vs. worker-owned organizations,

Human Relations, Vol 34, No 12, pp. 1013-1035.

Rice, A.K. (1958) Productivity and Social Organization: The Ahmedabad Experiment.

London: Tavistock.

Rosen, N. (1969) Leadership Change and Work-Group Dynamics. Ithica, New York:

Cornell Press.

Ross, J.E. (1993) Total Quality Management: Text, Cases and Readings. Delray Beach,

FL: St. Lucie Press

399

Ross, J.E. (1994) Total Quality Management: Text, Cases and Readings. (2nd edition).

London: Kogan Page.

Rossi, P.H., and Freeman, H.E. (1993) Evaluation: A Systematic Approach (5th edition)

Newbury Park, CA: Sage

Roy, D. (1952) Quota restriction and goldbricking in a machine shop, American

Journal of Sociology, Vol 57, No 5, pp. 427-442.

Roy, D. (1969) Making out: a workers' counter-system of control of work situation and

relationships, in T. Burns (Ed.), Industrial Man. Harmondsworth: Penguin.

Sashkin, M., and Kisher, K.J. (1993) Total Quality Management. San Francisco:

Berrett-Kochler.

Schein, E. (1979) Personal change through interpersonal relationships, in W. Bennis, J.

Van Maanen, E. Schein, and F. Steele (Eds.), Essays in interpersonal dynamics.

Homewood, I.L.: The Dorsey Press. pp. 129-162.

Schein, E. (1985) Organizational Culture and Leadership. San Francisco: Jossey-Bass.

Schneider, B. (1975) Organizational climates: An essay, Personnel Psychology, Vol 28,

pp. 447-479.

Schneider, B (1983a) Interactional psychology and oganizational behaviour, in L.L.

Cummings and B.M. Staw (Eds.), Research in Organizational Behaviour, (Vol 5)

Greenwich, C.T. : JAI Press. pp. 1-31.

400

Schneider, B. (1983b) Work climates: an interactionist perspective,in N.W. Feimer and

E.S. Geller (Eds.), Environmental Psychology: Directions and Perspectives, New York:

Praeger. pp. 106-128.

Schneider, B., and Reichers, A.E. (1983) On the etiology of climates, Personnel

Psychology, Vol 36, pp. 19-39.

Schriesheim, C.A. (1979) The similarity of individual directed and group directed

leader behaviour descriptions, Academy of Management Journal, Vol 22, No 2,

pp. 345-355.

Schriesheim, C.A., and Kerr, S. (1977) Theories and measures of leadership: A critical

appraisal of current and future directions, in J.G. Hunt and L.L. Larson (Eds.),

Leadership: The Cutting Edge. Cardondale, IL: Southern Ilinois University Press.

pp. 9-45.

Schulberg, H.C. and Baker, F. (1971) Program evaluation models and the

implementation of research findings, in F.G Caro (Ed.), Readings in Evaluation

Research. New York: Russell Sage Foundation. pp. 72-80.

Scott, S.G., and Bruce, R.A. (1994) Determinants of innovative behavior: A path model

of individual innovation in the workplace, Academy of Management Journal, Vol 37,

No 3, pp. 580-607.

Sewell, G. and Wilkinson, B. (1992) Empowerment or Emasculation: Shopfloor

surveillence in a Total Quality Organization, in P. Blyton and P. Turnbull (Eds.),

Reassessing Human Resource Management. London: Sage. pp. 97-115.

Sheppard, H.L., and Herrick, N.Q. (1972) Where have all the Robots gone? New York

Free Press.

401

Simmons, D.E, Shadur, M.A. and Preston, A.P. (1995) Integrating TQM and HRM,

Employee Relations, Vol 17, No 3, pp. 75-86.

Sitkin, S. B., Sutcliffe, K. M., and Schroeder, R. G. (1994) Distinguishing control from

learning in Total Quality Management: A contingency perspective, Academy of


Slater, R.H. (1991) Integrated process management: A quality model, Quality Progress,

April, pp. 69-73.

Snape, E., Wilkinson, A., Marchington, M. and Redman, T. (1995) Managing human

resources for TQM: Possibilities and pitfalls, Employee Relations, Vol 17, No 3, pp.42-

51.

Soin, S. (1992) Total Quality Control Essentials: Key Elements, Methodologies, and

Managing for Success. New York: McGraw-Hill.

Spector, P.E. (1985) Higher-order need strength as a moderator of the job scope -

employee outcome relationship : A meta analysis, Journal of Occupational Psychology,

Vol 58, pp.119-127.

Spencer, B.A. (1994) Models of organization and Total Quality Management: A

comparison and critical evaluation, Academy of Management Review, Vol 19, No 3, pp.

446-471.

Staw, B.M., Bell, N.E., and Clausen, J.A. (1986) The dispositional approach to job

attitudes: A lifetime longitudinal test, Administrative Science Quarterly, Vol 31, No 1,

pp. 56-77.

402

Steel, R.P. and Jennings, K.R. (1992) Quality improvement technologies for the 90s:

New directions for research and theory, in W.A. Pasmore and R.W. Woodman (Eds.),

Research in Organizational Change and Development (Vol 6). Greenwich, CT: JAI

Press. pp. 1-36.

Steel, R.P. and Lloyd, R.F. (1988) Cognitive, affective, and behavioral outcomes of

participation in Quality Circles: Conceptual and empirical findings, The Journal of

Applied Behavioral Science, Vol 24, No 1, pp. 1 - 17.

Steel, R.P., Mento, A.J., DiIla, B.L., Ovalle, N.K., and Lloyd, R.F. (1985) Factors

influencing the success and failure of two Quality Circle programs, Journal of

Management, Vol 11, No 1, pp. 99-119.

Steel, R.P., Ovalle, N.K., and Lloyd, R.F. (1982) Quality, circles in the Department of

Defense: Some preliminary findings. Proceedings of the 24th Annual Conference of

the Military Testing Association. pp. 40-45.

Suchman, E. (1967) Evaluative Research. New York: Russell Sage Foundation.

Tannenbaum, S. I., Mathieu, J. E., Salas, E., and Cannon-Bowers, J.A. (1991) Meeting

Trainees' Expectations: The Influence of Training Fulfilment on the Development of

Commitment, Self-Efficacy, and Motivation, Journal of Applied Psychology, Vol 76,

No 6, pp. 759-769.

Tannenbaum, S.I., and Yukl, G. (1992) Training and development in work

organizations, Annual Review of Psychology, Vol 43, pp. 399-441.

Tedeschi, J., Hiester, D., and Gahagan, J.P. (1969) Trust and PD game, Journal of

Social Psychology, Vol 79, pp. 43 -50.

403

Thomas, K.W., and Velthouse, B.A. (1990) Cognitive elements of empowen -nent: An

"interpretive" model of intrinsic task motivation, Academy of Management Review,

Vol 15, No 4, pp. 666-681.

Trist, E. and Bamforth, K.W. (1951) Some social and psychological consequences of

the longwall method of coal setting, Human Relations, Vol 4, pp. 3-38.

Turnbull, P.J. (1986) The `Japanisation' of production and industrial relations at Lucas

Electrical, Industrial Relations Journal, Vol 17, pp.193-206.

Verrna, A., and McKersie, R.B. (1987) Employee involvement programs: The

implications of non-involvement by unions, Industrial and Labor Relations Review,

Vol 40, No 4, pp. 556-568.

Wagner, J. A. (1994) Participation's effects on performance and satisfaction: A

reconsideration of research evidence, Academy of Management Review, Vol 19, No 2,

pp .312-330.

Wagner, J.A. , and Gooding, R.Z. (1987a) Longitudinal trends in American research on

the outcomes of participation: Effects of societal issues, methods, propensities, and

social attitudes on published findings, Administrative Science Quarterly, Vol 32,

pp.241-262.

Wagner, J.A., and Gooding, R.Z. (1987b) Shared influence and organizational

behaviour: a meta-analysis of situational variables expected to moderate participation -

outcome relationships, Academy of Management Journal, Vol 30, No 3, pp. 524-541.

Waldman, D. A. (1994) The contributions of Total Quality Management to a theory of

work performance, Academy of Management Review, Vol 19, No 3, pp. 510-536.

404

Wall, T., and Lischeron, J. (1977) Worker Participation: A Critique of the Literature

and Some Fresh Evidence. London: Mc Graw-Hill.

Wall, T.D., Kemp, N.J., Jackson, P.R., and Clegg, C.W. (1986) Outcomes of

autonomous workgroups: a long-term field experiment, Academy of Management

Journal, Vol 29, No 2, pp. 280-304.

Walton, R.E. (1977) Work innovations at Topeka: After six years, Journal of Applied

Behavioral Science, Vol 13, pp. 422-433.

Walton, R.E. (1985) From control to commitment in the workplace, Harvard Business

Review, Vol 63, March-April, pp. 76-84.

Warr, P., Cook, J., and Wall, T. (1979) Scales for the measurement of some work

attitudes and aspects of psychological well-being, Journal of Occupational Psychology,

Vol 52, pp. 129-148.

Webb, J. (1995) Quality management and the management of quality, in A, Wilkinson

and H. Willmott (Eds.) Making Quality Critical: New Perspectives on Organizational

Change. London: Routledge. pp.105-126.

Whyte, J. and Witcher, B. (1992) The adoption of total quality management in Northern

England, Durham University Business School, Occasional Paper No. 9236.

Wickens, P.D. (1995) The Ascendant Organization: Combining Commitment and

Control for Long Term, Sustainable Business Success. Basingstoke, Hampshire:

Macmillan Press.

405

Wilkinson, A. (1992) The other side of quality: "soft" issues and the human resource

dimension, Total Quality Management, Vol 3, No 3, pp. 323-329.

Wilkinson, A. (1994) Managing Human Resources for Quality, in B. G. Dale (Ed.),

Managing Quality. 2nd edition. Hemel Hempstead: Prentice-Hall. pp.273-291.

Wilkinson, A., Allen P., and Snape, E. (1991) TQM and the management of labour,

Employee Relations, Vol 13, pp. 24-31.

Wilkinson, A., Marchington, M., Goodman, J. and Ackers, P. (1992) Total quality

management and employee involvement, Human Resource Management Journal, Vol

2, No 4, pp. 1-20.

Wilkinson, A., Redman, T. and Snape, E. (1993) Quality and the manager; An IM

report, Institute of Management, Corby.

Wilkinson, A., Snape, E., and Allen, P. (1990) Total customer service, The TQM

Magazine, October, pp. 291-294.

Wilkinson, A. and Willmoft, H. (1995) Introduction, in A. Wilkinson and H. Willmatt

(Eds.), Making Quality Critical: New Perspectives on Organizational Change.

London: Routledge. pp. 1-32.

Williams, A., Dobson, P. and Walters, M. (1993) Changing culture: New

Organizational Approaches. (2nd edition) London: IPM.

Willmott, H.C. (1992) Post-modernism and excellence: The de-differentiation of

economy and culture, Journal of Organizational Change Management, Vol 5, No 1,

pp.58-68.

406

Willmott, H. (1993) Strength is ignorance: slavery is freedom: managing culture in

modern organizations, Journal of Management Studies, Vol 30, No 4, pp. 515-552.

Wood, S. and Peccei, R. (1995) Does total quality management make a difference to

employee attitudes?, Employee Relations, Vol 17, No 3, pp. 52-62.

Yammarino, F.J., and Dubinsky, A.J. (1992) Superior-subordinate relationships: A

multiple levels of analysis approach, Human Relations, Vol 45, No 6, pp. 575-600.

Yin, R. (1984) Case Study Research: Design and Methods. Beverley Hills : Sage.

Yukl, G. A. (1989) Leadership in Organizations. 2nd edition. Englewood Cliffs, New

Jersey: Prentice Hall.

Zand, D. E. (1972) Trust and managerial problem solving, Administrative Science

Quarterly, Vol 17, pp. 229-239.

407

Documents

the impact of a tqm intervention on work attitudes: a longitudinal